Our Fabric Resolutions – Ep. 491

Mike and Tommy share their 2026 Fabric resolutions — from doubling down on Power BI embedding and data warehouses to validating AI-generated code and building custom workloads. Plus, Microsoft acquires Osmos to bring agentic AI to Fabric data engineering.

News & Announcements

Microsoft Acquires Osmos

Mike shares a surprising development: Osmos, a company that had been building an AI-powered data engineering workload for Fabric, first sent an email saying they were deprecating their workload — then weeks later, Microsoft announced the acquisition. Bogdan (one of Microsoft’s data engineering leaders) announced that Osmos will extend Fabric with agentic AI for data engineering.

Mike sees this as a smart strategy: smaller companies can innovate fast and build rich products on top of Fabric, then Microsoft acquires them rather than fighting through internal bureaucracy to build from scratch. Tommy agrees — this signals Fabric’s maturity. It follows the normal Microsoft product lifecycle: release a platform, let the ecosystem build on it, then acquire the best add-ons to make them seamless.

The only other notable Power BI-related acquisition they can recall? Data Zen in 2015 (with help from the live chat) and ProClarity back in 2006. This is the first Fabric-era acquisition, and likely not the last.

Main Discussion: 2026 Fabric Resolutions

Tommy sets the ground rule: resolutions should be measurable — not “I will build more notebooks” but something you can actually track. Here’s what Mike and Tommy are committing to in 2026:

Mike’s Resolution #1: Power BI Embedding Education

Mike’s seeing a surge of interest in embedding Power BI into existing applications. His company’s product, Entexos, has been an embedded solution for 6-7 years, but most people still think embedding means read-only reports. The reality is much richer:

• Edit mode — users can modify reports within the embedded experience
• Create new artifacts — build new reports and visuals directly
• Semantic model interaction — add measures, build custom visuals on top of the model
• Cost optimization — pay less per user than $10-14/user Pro or PPU licensing

The resolution: ramp up education through training.tips, showing organizations the full power of embedded experiences so they stop leaving value on the table.

Tommy’s Resolution #1: Validate AI-Generated Code

Tommy’s been relying heavily on Copilot, ChatGPT, and Claude to write Python and Spark code in notebooks. It’s been productive — but he’s asking himself: “If OpenAI retired tomorrow, what would I do?”

The resolution: spend more time understanding and validating AI-generated code rather than just trusting it works. Not abandoning AI — but making sure he can read, debug, and verify everything it produces. His practical tip: always prompt with “ensure you’re adding comments like I’m a dummy” so every code block is self-documenting.

Mike’s approach is different — he leans on agents to explain code line-by-line rather than learning the syntax himself. The ability to take any code and say “explain this to me line by line” has fundamentally shifted how he works.

Mike’s Resolution #2: Make AI Actually Save Time

Mike’s honest assessment: AI is impressive and helpful, but it doesn’t save him time yet. There’s too much back-and-forth — giving requirements, reviewing output, rebuilding when the AI’s approach is inefficient. It’s like working with a junior developer who delivers the right result but not the way you’d build it.

His specific frustration: watching YouTube creators build beautiful, highly-styled Power BI visuals with dynamic text, colors, and conditional formatting — and wanting AI to handle the 15-16 clicks and settings to reproduce that visual without manually learning every menu.

The vision: “supplemental agents” that handle specific micro-tasks (style a button, build a color palette, configure a visual) rather than trying to build entire reports. Tommy coins the term: 2026 is the year of the supplemental agent.

Mike also floats the automation dream for the podcast itself: record → auto-generate blog post → transcript → social clips. The pieces are there, but the automation setup still takes too much effort. (2026 might be the year it tips.)

Tommy’s Resolution #2: Embrace the Data Warehouse

Brad from Microsoft sold Tommy on it. The data warehouse in Fabric is getting consistent, substantial updates — not flashy, not AI-powered greetings, but the foundational admin and data processing capabilities that actually matter.

Tommy’s framework: “What does this do better than the lakehouse? It has to accomplish X better for me to use it.” And he’s seeing more cases where the warehouse wins. The resolution: not just build warehouses, but build them the right way and move more data endpoints from lakehouse-only to warehouse.

Mike extends this further: Fabric shouldn’t just be the analytics destination — it could become the transactional system too. SQL databases, Cosmos DB, and upcoming database options in Fabric could serve small-to-medium applications directly. The key blocker: autoscaling. You can’t run production applications on a platform that throttles queries when you hit capacity limits. Pay-per-use pricing (like Spark’s autoscaling) needs to extend to all database workloads.

Tommy’s Resolution #3: The Intelligence Developer

Tommy’s leaning into the “intelligence developer” concept from Episode 487. His role has evolved from “data engineer who learns all the tools” to “conductor who understands how data flows through an organization.” The focus areas:

• Data governance and trust — AI agents are only as good as the data underneath them
• People and process — the gray area you won’t find in documentation
• Workflow orchestration — understanding the full pipeline, not just individual tools

His conviction: the only way you trust AI is if you trust the data. The only way you trust the data is if you have good data maturity. That’s the intelligence developer’s job.

Mike’s Resolution #3: Build More Fabric Workloads

Mike’s company is investing heavily in custom Fabric workloads. He sees workloads as the corrected version of what custom visuals tried to be — a platform for external developers to build on top of Fabric.

Where custom visuals went wrong: unclear licensing, difficult development experience, and absurd pricing ($10 per visual per user). Workloads fix this with better API access, proper user permission integration, and a more developer-friendly ecosystem.

The plan: build fast, ship often, see what sticks. Some will stay in preview, some will evolve — taking a page from Microsoft’s own playbook.

Bonus: Tommy Gets Personal

Tommy acknowledges listener feedback directly: “We hear you. Sometimes people are like, ‘Is he alive on the podcast?’” His personal resolution: bring more energy, be more involved, and lean harder into the intelligence developer identity on the show.

Episode Transcript

Full verbatim transcript — click any timestamp to jump to that moment:

0:17 [music] Good morning and welcome back to the explicit measures podcast. We are back live again on the episodes. We are going through our New Year’s resolutions this year. What’s what’s our New Year’s fabric resolutions? I guess it’s 2026, man. We made it another year. I personally have never been more excited for a year than this coming year. I don’t know why. I I feel good

0:50 About it. I already have some things planned up. So, it’s one of those things where you’re like, I I could not wait before Christmas break for the next year to start. Really? I don’t I don’t not sure if I had that same sentiment. I was just plotting along, doing things. I do feel like though whenever I go out on break, I work hard and work a lot [laughter] all throughout the weeks. But when I take a break, the first couple days of break are like I’m still thinking about work. I try to get work things off my plate. At some point, I’m able to push enough work off of my plate that I’m not thinking about it. I’m

1:23 Actually able to take a break. And then at the end of my break, I’m like, “Oh, I got to go back to work. I got to switch gears again.” And that’s that’s very difficult. Not for me. No. So, because by the end, and here’s the thing. So, the week of Christmas, I I worked the Monday and Tuesday before Christmas Eve. And my wife’s like, “Well, why are you putting so much work in?” because I I’m like I had I have a big thing coming up this month which I’m really excited about a big project. I’m like I need to get out of my head because it’s just going to be in my head unless I get to a point you’ll dwell on it. You’ll keep

1:56 You dwell on it, right? Yeah. It’s always going to be in my head and that’s work in general for me. But I like that. But that being said, after two weeks of kids being home and it was a week and a half I took off, you start feeling the little the twitch and that for me I’m like it’s quiet down here, ? And what? It’s just your voice. It’s not someone putting their nose in someone else, their finger in someone else’s nose or the the weird things. Granted, I miss my kids, but this is good. This is very good. So, yeah, a lot of my day is just spent thinking and planning and working with

2:29 People on computers. I don’t have a ton of time. It’s very different for my pattern to have be upstairs with all the kids and all the extra noise and the things that they’re doing. They’re always constantly trying to be busy. I’m like, I just like to sit still every so often because that’s what I do every day. Anyways, all right, good. , any New Year’s traditions that you do, Tommy? Yeah, so actually I think I also brought up the reason I was excited for this year. New Year’s is actually, I think, my favorite holiday. It’s my favorite get together with people. And let me try

3:02 To sell you on this because I know this is for a lot of people, they’ve never even heard someone say this before. Sure. So, when you think about every holiday, usually like Thanksgiving, Christmas, the big ones, it’s family or, usually family. And it’s it’s good, but it’s a little stressful because if you’re hosting, let’s say you’re hosting, well, you got to make sure the food’s ready and you got to make sure everyone’s fed and where they’re staying. So, there’s a little added stress to the holiday. Also, most holidays you are also having them extended stays. So it’s not just the single day, which is stressful,

3:36 Especially if you’re hosting. Even if you’re not hosting, you don’t want to impede on someone else’s traditions. You just are there. So there’s all that little extra added work that’s going in even though you’re just trying to relax. New Year’s Eve, man, the only thing you’re supposed to do is just enjoy everyone’s company. There’s no expected traditions or like, “Oh, I can’t believe they didn’t bring out cookies for Christmas.” you just there’s no conversation you’re having in the wife in the car like, “Can you believe that they did X, Y, and Z? There was no turkey. Oh my gosh, that doesn’t happen. The best thing is when New

4:10 Year’s finally hits, all it is is people just going, “Happy New Year’s. There’s hugs. It’s people that you enjoy, hopefully that you’re with, and it’s just a nice easy time. Not saying the other holidays are bad. I’m just saying this is the only holiday without any stress.” I can say the preparation is be a little bit less. Just get some balloons, get a couple things, watch something on TV of the ball dropping and then just relax. We have spanicopa or cheese sticks or chicken wings or whatever it is, something people pick, they can do it themselves. So, it’s great. And the

4:43 Tradition that we have is actually from my best man, my best friend, and it’s a I think it’s a Cuban tradition. He told me that. Every once New Year’s hit, everyone gets a bag of 12 grapes to signify every month of the year. And what we did or what the tradition was is you throw it up in the air, you catch it. Every grape you catch is a good month coming up. Oh, okay. So, I’m happy to I’m happy to report it’s I’m going to start strong. The summer’s going to be a little lean. [laughter]

5:16 June and July are not looking great. And then and then we’re going to ease we’re going to have a good end of the year, but we’re going to have a great first quarter. But I’ve always like that. Yeah. So just a Exactly. fun. No stress. So that’s something I really enjoy doing. We this year for our family, we took a trip this year. And so while we were on our trip, we were gone during New Year’s. So that was when we were able to we typically would let the kids stay up late till midnight if we were at our house, but they’ve been getting older and so they’ve been able to more easily

5:49 Handle up late nights. I don’t know how old your kids are, Tommy, but the younger ones don’t seem to make it all the way till like midnight. So they want to in the past what we’ve done is we’ve done like, TV shows or like there’s like early Netflix had some like pre-show that was like, earlier in the afternoon. So this time we did a late night. Everyone stayed up and we watched the ball drop and we had balloons coming down from the the ceiling and everyone was partying and having fun together. This was the first time they’ve actually had like a real like go somewhere and and be at a place and actually participate in New

6:20 Year’s which is a lot of fun. So awesome. I think that was a good making core memories I guess. Core memories. Well that actually leads to a really great question Mike road trips with the family. You as the dad how do you handle that? And I’m going to preface this with the management of let’s say it’s a 6 to n hour ride. I don’t know how long your road trips are. How long are they usually where we we do a lot less road trips anymore. Okay. Most of the stuff that we need to get to are flying trips and now that our kids are older, we can do flying trips much easier.

6:52 I definitely also notice well again we have a minivan so we’re we’re totally a minivan family and everyone can like get in and zone out. Our kids are now old enough that most of them have devices that they play on. So, we pretty much shove them in the back of the car, let them have free reign of devices, and we just haul. I like driving. I don’t mind longer trips. I can do pretty good drives. Although, as I’ve been getting older, I’m just more anxious about getting there. So, a lot more of what we’ve been doing is the longest drive we have is like 30 minutes the airport and then we travel together to wherever we’re going.

7:24 I gota on an airplane. We’ve been doing a lot of Florida type trips. Florida trips. That’s nice. The place where Tommy came from. Yeah. Thank you for that terrible reminder. [laughter] but so we actually surprised my sister. She turned 40 and we’re she lives in Alabama. And I have to preface this the Did that the place with the highest per capita of people with PhDs is actually in Alabama because of the the rockets. It’s yeah,

7:55 I feel like there was a TV show for a while that was like Alabama. It was Redneck Science or I can’t remember what it was. It was something It was something about like Alabama scientists and there were just a bunch of PhD guys sitting around doing crazy building interesting things down in Alabama. But yes, there’s a lot of really smart people all around the rocket space. I have to preface that because every time I say I went to Alabama for Christmas, they get a turn of the head like why? but no, if you’re from Alabama, it’s a great place. But we went for New Year’s and the drive is nine hours.

8:27 Yeah. And I have a, eight, six, and 5year-old or eight, jeez, , nine, seven, and 5year-old in the car. And, but my goal is, look, you’re sleeping in the beginning because we leave at like 4:3,5 in the morning. I do not like driving when it’s like coming the evening. I like just getting there and then being able to relax. So, I’m a big proponent of leaving early. Now, you you’re already up early anyways, just because your mind clicks on and you’re up and running. Yeah. Does that mean you go to bed earlier then? Would you go to bed earlierish?

8:59 No, I we the kids do, but for me like my wife is usually she’s actually the packer planner and that’s a great and people like you don’t pack. I’m like ask her ask her if she wants me to pack. That’s her thing. So I am I am the luggage picker upper basically. So I am putting it in the car doing the Tetris with it. But she’s doing all the organization with that. So we do that the night before. So the morning we wake them up, we go, they sleep. I just rock out to Billy Joel while everyone sleeps in the car for the first three hours. Yeah. So I got there’s a few nice live albums,

9:32 But Yeah. Because then you’ve got like you’re out of Chicago. Like you’re out of the traffic area. There’s not a lot of You’re not sitting in traffic 30 minutes from your house. Those things. Yeah, that would be horrible. Just like we left earlier to go to Alabama, so we did not hit the Chicago traffic at the rush hour. Yes. This is what dad should do. I I think and but we only stopped for gas. If there’s like You heard of Bies? Yeah, I heard of Bies. Yeah, I’ve never been to one, but but my wife and my daughter just recently visited one and was like, “Wow, this place is amazing.” I’ve never been there yet. So, we’re to the Wise. So, for those who are

10:06 Not aware what Bies is, it’s the best gas station you’ll ever go to in your life. It’s a giant it’s almost like an amusement [laughter] park and there’s they’re serving meat and brisket and you could buy almost anything in there and there’s like probably 70 gas stations. The bathrooms are what they sell is actually the big selling point. You will never go to a a nicer bathroom in your life and that includes in your own house and and that’s one of the big selling points. However, word to the wise Mike go early. When we went, we got because it was on, it’s closer to

10:41 Alabama. We got there around two o’clock. It was like an ant farm. I’ve never seen more people in a condensed area. And this is someone who originally is from New York and lives in Chicago and I’ve never seen more people crammed into one place at one time. So, wow. I’ve never been there. Yeah. I couldn’t speak specifically to that. When you do the flights though, do do you have a a set pattern of we get in, everyone just takes a seat or do we eat at the airport? Like what are some of the tips that you have on flying with family?

11:13 Well, since I’ve been traveling a lot to Europe recently for speaking engagements, I’ve been able to get status on Delta, which has been a super help. And with that, I’ve been able to push myself to a higherend membership to get into the lounges. So now my kids have been royally spoiled because we go to the airport and they’re like, “Quick, where’s the Delta Lounge?” And so like they all they want to go there. Yeah. They go there and they sit and they just hang out, eat food, and get, lemonade, unlimited lemonade for themselves. And they’re like, “Flying’s great.” I’m like, “Yeah, it is once once you get to like the nicer end

11:47 Of things.” So that’s pretty much our our route. We we like to fly direct, but in most places where we go, especially when we’re going to Florida, Delta doesn’t have a lot of direct flights directly to Florida. So, we spend a little bit of time in Atlanta, and the kids are always like, “Okay, which lounge are we trying out this time in Atlanta?” Like, we’re always trying out different lounges there. Yeah. It’s it’s like a cheat code. So, they absolutely love that. have a lot of fun there. So, that’s like our routine. We get we get up, we pack our stuff. I don’t like to leave early in the morning. I’m not an early morning person. My wife and and is, so she’s like, “Let’s get up earlier.” here. I’m

12:20 Like, I just don’t and I’m grumpy the rest of the day. So, , we try to leave, midday to afternoon on the flights and then get there when we get there and and and hang out when we get Yeah, we’re different that way. If I fly, Mike, I’m trying to get breakfast at where I land. That’s my goal. Like, let’s get in time. So, breakfast is still being served. That is what I want. So, you’re like, listen, if we hit dinner, great. [laughter] We’ll get there when we get there. So anyways, amazing. All good. Well, all these New Year’s traditions, let’s transition into a

12:52 Little bit more of like so, we’re talking about New Year’s traditions or New Year’s things that we’re going to do that that we have over the new year. Let’s poke it a bit here of what things would be traditional or what things we want to do for this next year of 2026 around fabric. Where do we see fabric? what are we going to do specifically? What are we going to work on inside fabric to become better at to get more skilled on? where do we see ourselves being more equipped or what are we going to work on in the year of 2026 around things inside fabric?

13:26 I like this too. And Mike, what I would like to do with you with this is for the person who’s giving their resolution, does the other person agree? Is that something they’re going to add to their resolution list or is that something they feel that’s not a need? And yeah, I I like this as well because one thing I’m going to do, Mike, is with any resolution. I I watched this incredible video on how to actually set resolutions. And this is how I’m going to try to frame a lot of what I’m doing rather than saying I will do, I’m going to run, right? You want to say I’m going to run three times a week. So,

13:59 It’s almost something you can measure. So, that is how I’m going to frame a lot of my resolutions here rather than I am going to build more notebooks. that doesn’t usually happen, especially over a long period of time. So, Mike, yeah, I think this is all stuff that’s just going to make us better at what we do or organizations, things that we’ve just seen that come up. And Mike, do you want to start? Do you want me to kick it off? How do you want to do this? I can start if you’d like. Sure. So one trend or pattern that I’ve been seeing particularly with our customers today and and the customers that I’m working with more and more now is I’m

14:32 Seeing a whole lot more interest around PowerBI embedding embedding experiences embedding PowerBI experiences inside their existing applications. We’ve been doing a lot more help around this. there’s a lot more interesting things coming out with AI now. and so I think for me around a 2026 resolution will be is figuring out more places where we could strategically place PowerBI embedding. we’ve got a really good product that we’ve been building for the last about six seven years or so. And because of that we’re going to

15:05 Continue to invest in building a really solid product. So if you haven’t heard we have this product called Entexos. Intexos is an embedded solution for companies to start embedding their PowerBI reports into their own application if they want their own portal for reporting things around PowerBI. I think people only think embedding is just read only. It’s actually not that. It’s actually has a lot more capabilities. , and also now with a the whole bunch of APIs that you get from Microsoft, you can actually build embedded experiences where you’re editing the model or adding your own measures or doing other adding other visuals experiences, not standard

15:40 PowerBI visuals, but using the model as a baseline and then building other visuals and things on top of it. So, I think for us, what we’re going to really spend a lot of time on this year is going to be emphasizing and focusing on how rich of an embedding experience you can get through PowerBI. You can not only just read reports, but you can read and edit. You can create new items. You can create new artifacts using embedding. So, this really can be a foundational element for your data solutions inside your existing applications. I’ve seen this for many years. Maybe I haven’t been as vocal about it.

16:13 But if nothing else, I’m going to be much more vocal around embedding experiences. I’m going to do a lot more education. , one thing you’ll notice for me in the new year, this year, 2026, is I’ve been running a tra a class called training.tips. And so, I’m going to go back to basics a little bit. , we’re going to start doing more content on training.tips. And I want to do a lot around what is embedding, how do you use it, how do you get started with it, , what solutions already out on the market that you could you could use out of the box so you don’t have to

16:45 Build everything from scratch. So we’re going to focus a lot of our attention this year on this more enriched experience or enriched space around embedding. So So give me the statement then for the resolution. Is it I will get more clients or more customers with an embedding solution or is it I’m going to improve the product and and make a complete solution here. Yeah. So I think I’m going to really point out is I’m going to continue to educate or spend more time educating around the value of PowerBI. So I think that’s really where we’re going to spend

17:18 A lot of time. I think if you understood really the implications of like how it works, how to get started with it and solutions that are out there. I think if you just understood I think more organizations would be more willing to start with PowerBI betting because it is very robust and it handles a lot of your if you think about the users in your organization the ones that are consuming and the ones that are creating embedded does a really good job around consuming reports and creating reports. If you’re editing semantic models, that’s more of a powerbi.com

17:50 Experience, but the embedded solution allows for more flexibility there. And it’s also, to be frankly honest, it’s a cost optimizer. Yeah, it helps you optimize costs. You don’t have to pay $10 or $14 per user. you can pay much less per user to implement an embedded type solution. So, it’s actually a really good cost savings as well. So, I think there’s a lot of considerations to think about it. I’m just going to educate more people about it so that they’re more willing to use it and and jump in the in in the boat with embedding. Yeah. Not going to pander here, but I I I constantly bring people when they ask

18:24 For an embedding solution, I go, “There’s something you should check out.” For me, I’m not probably going to make that a major focus. And this is actually something I love, Mike, about doing the podcast with you. When we started, my wife actually said, “Aren’t you competing with Mike?” Because you’re going to consult into. I said, “It’s like two comedians doing a podcast. There are different styles. There’s also different products here too. Yeah, there’s some overlap, I’m sure. Sure. But there’s a ton of things like I’m never going to need to touch embedded. Why? You have a great solution. Why would I go into something that I use or recommend to my clients too? So, no, I I

18:58 Like that. So, for me, yeah, if I have people who are talking about it, I would point you to you. There is nothing better I’ve seen in the market. I’m personally not going to put that on my list. But I will add on to what you’re saying though, one thing that you said. One of my things I’m going to side note on resolution is going to be you’re going to see my face more hopefully online on on training. We’ve talked about doing some more training programs. There’s a lot of ideas I have too. So that’s something I have. But Mike, I like your resolution here. So I’m going to start with a fun one here around fabric.

19:30 Sure. I am going to learn and validate my own code coming from AI, especially in notebooks. too often. Mike, I’ve been relying and it’s been great and I’ve gotten a lot done. I’ve been able to rely on Copilot, Chatgpt, Claude to help really build out my solutions for a lot of the Python and Spark. But if I looked at the code myself, if all of a sudden, OpenAI said, “We’re done. We retire, we’re out, what would I do?” So, I’m gonna try to be as self-sufficient

20:03 As I can with the code as well. Because here’s the thing, Mike. It’s great that I can sustain even help clients using AI, but there’s still that head knowledge that I I still think and maybe this is like a guy who read a book and said, “I don’t want Google. You need to read a book.” Maybe it sounds like that, but to me, I still think it’s important that I’m going to be able to validate the code that is getting generated. And I’m going to spend more time on that rather than being lazy on, oh, that’s what Chat GPT said. It looks like it looks like it works. and I’m just going to go run with that. So, that is gonna

20:36 Be something that’s going to be part of my process now when building out solutions or more often. That’s interesting. , are you going to take a are you thinking about taking like a class specifically like an online class around something that’s, Python based or or thinking through that? I’ve done that. It’s going to be more on honestly what I did with PowerBI. It’s the Googles, man. It’s the searching. It’s looking more on the forums. It’s testing it out myself and trying to write it. the reverse engineering the way that I’ve learned, the way that I’ve continued to learn and using that

21:08 Approach with my notebooks. Okay. Excellent. Yeah, I think I think Python’s a very neat general language. It’s very easy to learn. I’ve really enjoyed it. I’m a huge proponent of learning by watching something else code and produce the code for me. Not me at all, which is well well I I like watching agents and and telling an agent, hey, I’m trying to do this. Can you please repeat it? Right. And I’m I’m finding reasonable success with agents going from I’ve got this, SQL cell block. Can you rewrite it in Python?

21:41 Yeah. Oh, speaking of which, I believe there was an interesting announcement. So I’m thinking about AI and data engineering and things as well. I believe I So do you use any of the workloads, Tommy? Have you played with any of the workloads inside fabric? Yours. , and then there’s a few others in there that I play I played with, but nothing that is part of a daily workflow. So, definitely, I touch what I can. Okay. The reason I’m bringing this up is because I think you’re going to bring up one of the workloads was announced

22:13 Recently and one of the workloads, well, not recently. Let me say this again. You told me earlier last year, one of the workloads is called Osmos. That’s it. and and Osmos is an AI generated like an agentic way of like using AI and data to build out an AI based solution Microsoft and I got an email probably about I I tried their workload. I had the workload running and I got an email that was interesting. It said Osmosis is is basically going to be deprecating. They’re not going to support the workload anymore. The workload is going

22:45 Away. It’s not going to be here. And then another week later or two weeks later, I got a notice on from Microsoft. Microsoft announces the acquisition of Osmos to accelerate autonomous data engineering in fabric. So Bogdan who is one of the leaders of the data engineering space has just recently announced that they’re acquiring this Osmos company. Now I played with the company. It was interesting. the link to the article is actually in the chat in case you wanted to check it out. But Microsoft has announced that they’ve acquired them and Microsoft and Osmos is going to extend fabric with Agentic AI

23:19 For data engineering which I find this to be very interesting. a let me say it this way. I think Microsoft is wise to spend time acquiring smaller companies that are building really rich products on top of their existing stuff because these smaller companies can innovate quickly, build out good stuff, create things that are useful to people, and then Microsoft doesn’t have to go through all the bureaucracy of trying to build a product and security things, all these other things. So I think it it makes a lot of sense for Microsoft to acquire companies that are building

23:51 Workloads so they can get the workloads that Microsoft needs done quickly but yet these independent companies can work and build them faster for them. So this is interesting. This is a trend I have not seen before. Microsoft has very rarely acquired things. I think the last acquisition around PowerBI was there was some dashboarding tool a long time ago. Tommy this is the very early days of PowerBI. SharePoint dashboards in SharePoint or a little after that? It was a little after that. It was the It was the company that was doing like cards and interesting cards. Remember that way back in the beginning?

24:24 Yeah. Yeah. That wasn’t micro. Well, there so there’s the power KPIs, but that was a Microsoft visual. Yeah. But there was another company that they acquired at the time very early on. And I can’t I’m the name is escaping me and Alex Powers is listening to us right now cuz apparently he’s a loser like we are and both working at on a Tuesday after [laughter] after break. So maybe Alex knows what it is and what that company was. But there was an a company that was acquired very early on in the PowerBI days and but since then nothing has been acquired. There’s been no acquisitions of companies to build

24:57 Anything around PowerBI. So this is the first thing I’ve seen that Microsoft is acquiring around the fabric data engineering space. Microsoft pretty much builds their own stuff. Yeah. So this is interesting to see this happening. This doesn’t happen very often. This is the normal this is the normal cycle too because you release a product and you make it open source people start actually building viable solutions and Microsoft goes can we make this seamless? This is the this is the normal strategy. So now we’re getting this is also showing the maturity of where fabric’s going to. And Alex, I guess if we do have someone from Microsoft

25:28 Listening, you want to buy a podcast. How [laughter] about that? So no, but this is No, I don’t think they buy podcast at this point. I don’t think they buy podcast. , we’ll go on the road. I’m happy to go on the road with you. , but no, I think this is the maturity of where fabric’s going because this is the normal thing you see with most Microsoft products. People make add-ons. Once that product is actually got to a point where those add-ons are pretty useful, Microsoft, well, like we can make this seamless with what we’re doing already, this is already a complete solution. They buy it. So, this is already this is a good

26:02 Sign for fabric to me. Yes. So, and along this lines, so Alex, I think you’re right. You’re spot on. So, Alex is listening. Alex Powers, thank you very much for doing that. I believe Proclarity is the company I’m thinking of. They had like this like dashboarding KPI card-based experience and Microsoft bought them a long time ago and this was so Microsoft bought PowerBI in 2000 or came out with PowerBI basically in 20 2005. Can you believe it Tommy? It’s been almost 2005 15. That’s 20 years. Oh 20 years. Sorry. Yeah. This this is not the one I’m thinking of then because

26:35 This is this was acquired in 2006. So Microsoft acquired ProClarity in 2006. Wow. You’re going really No, I that’s not the one I’m thinking of. I think there’s another one that I was thinking of. I can’t I’m It’s escaping me now. I can high school in 2006. I can literally see the cards, the KPI cards that was being built there. I I’ll Google it later and I’ll probably bring it up in at a future episode after I go find what the actual company was. Yeah. So, [laughter] anyways, good acquisition there. Anyways, moving

27:07 On. so you you did your thing Tommy around more Python learning code doing those things as well validating all and not just Python whatever the code I’m building whether it’s also the warehouse or it’s in a lakehouse I’m going to be validating anything that comes out of AI and spending more time on that. So Mike is that something you think you’re going to take with you because you’re pretty proficient already but still I’m assuming a lot of what you do is still generating from AI too. Yes and no. I would say more more of our experience. So, I do a lot more of like the architecting side of

27:41 Things. I’m doing a lot more of the business acquisition side of things in my company. I have a team of people behind me that do a lot more of the data engineering, the the the work day-to-day. , one thing that we’ve been finding extremely valuable for us has been using AI to help write code. So, again, we’re still going to use that a lot. Do I need to actually learn the Python? Probably not. if anything else, I’m going to be more engaged in using agents to help me understand what Python easier has been written or writing more Python. A lot of times though I’ll ask it directly straight up here’s some Python what is it doing

28:14 Line by line tell me what this is doing and it’s really good talk this is something that has really shifted my workload like just the idea that I can take any code anywhere and give it to some agent and say explain this to me line by line what’s happening I mechanically I can understand what the code needs to do I don’t necessarily know all the individual functions what they mean like meld in Python yeah Yeah. Yeah. Like I may not understand it, but there’s an expert at my disposal that costs me almost nothing to send

28:46 Information to it. Like it that’s the part that’s really shifting things here for me. what I do, Mike? Anytime I actually generate code, I started this probably two months ago, is every time I ask it to generate code, I always ask, ensure you’re adding comments like I’m a dummy. Add comments to the code explaining what that section or that definition or that line is doing. And I said, “Whatever that block of code is, please add comments like I’m a dummy to for other users as well, but also for me.” Yeah. All right, Mike, what do you got for resolution number two?

29:21 , let’s go after number two here. So, my my number two resolution is maybe this is a bit more, ideiation or or or it still counts. Ethereal, I guess. Maybe it’s a bit more like fuzzy. It’s fuzzy. It’s not as tangible. I don’t know. [laughter] Use bigger words that Tommy doesn’t understand. And we’re done then. The revolution you completed. Completed. You accomplished it. Yeah. I’m I’m still at the point, Tommy, where AI is really valuable. I I find it to be

29:57 There there are moments in time it very much impresses me. It really does things that I can’t do on my own. But I often feel like AI is not quite polished enough to actually take less time of what I’ve been doing. Let me let me restate this another way. Maybe inside the the things that I build in fabric, a data flow, a pipeline, a notebook, I can talk to the AI and it can do a lot of things very helpful for me, but I feel like I spend a lot of time talking to it and giving it requirements and building things for it. And there’s a lot of back and forth between me and it to get it to

30:31 Develop or build what I want inside fabric at this point. I don’t think AI is at a place right now where I can give it less direction and have it get more of what I want done. Right? So less direction, more that I want done. Okay? So think of it, think of it like this. I got a junior developer on your team. I can give you specific instructions. Hey, I want you to build a data pipeline that does this, right? and the AI goes out and does something or the junior developer goes out and builds something in a pipeline and they come back to me with the result and it does what it it

31:02 Does what I expect it to do. It it builds the right pipeline but it doesn’t quite do it how I would do it, right? It’s it’s a little bit well that’s a little inefficient here and you’re copying the data two or three times. That’s not really very efficient. So what I feel like a lot of times is I had to like come back and rebuild a couple times. Okay, I’m going to come back to my thing. The chat has has come up with the answer. Chat, you’re amazing as always. Data Zen, I [clears throat] believe Data Zen is what I was thinking about. Acquired in 2015 by Microsoft.

31:35 Going back to my earlier conversation. Oh. Oh. Oh wow. Okay. So, thanks chat for coming up with that. that that is exactly what I was thinking about was Data Zen was acquired by that’s the only other acquisition I’ve seen around PowerBI and fabric so far. Now, Osmosis. Okay. Anyways, back to back to my story. Sorry. , is there an AI thing I didn’t hear about, [laughter] man? No, no, no, no. That that was the end. Thank you. , , chat for figuring that one out. Okay. Anyways, back over to incorporating more AI in my workflows. I’m with you. I’m going to keep learning about AI. I’m

32:08 Going to keep teaching about things that I’m figuring out where AI works because I think this has the potential for really making my workloads more efficient. I’m seeing a lot of really interesting things around people building like reports with AI. being able to give it the spec. There’s a lot of MCP servers are pretty interesting. Using AI to push things around inside a semantic model I think is very effective, but it doesn’t quite save me time yet. I still feel like I’m in the learning mode where I’m still doing too much talking to the AI. Maybe I don’t have the right instructions. Maybe I don’t have the right,

32:40 I haven’t quite worked out my situation. And this is this goes maybe even beyond fabric at this point, but like even automation of the podcast. One thing that Tommy and I have been thinking about doing here for a long time is for five years. Well, I want to take we have we’ve been doing all these big pieces of content. It’s an hourong episode. Why can’t I get this automated from hey, I want to record the video with you, Tommy. I want to have it broken down from the video into a blog. I want to have it broken down from there into like a transcript. I want to have a a blog post created automatically with the like there’s all

33:12 These things I want to automate and right now the automation is just it feels like it’s so much work to get those automation things started yet with AI that we’re I’m not quite there yet and I think 2026 has the has the ability to be the year of where the AI starts tipping the scale and the AI becomes good enough to actually really start feeling like it’s integrated well enough it’s actually saving me time and I don’t have to give it so much extra instructions. I can be a little bit less verbose and it gives me better

33:47 Results back. Right. So, I think use of words again, but yeah, I’m listening. Does that make sense? Like what I’m trying to say there. Yeah. So, and it’s effective but not as effective as I want it to be. So, I’m going to continue learning on it. I have some comments here. First, I thought you were going to say about the podcast. I’m gonna basically take the good parts of Tommy’s area and when he’s not tired and just make that an AI voice chat during the podcast and then you just cut me right out. , [laughter] we’ll get the parts where Tommy’s had coffee and we’ll make we’ll make Tommy a GPT. , no, but what I I’m going to push back a little because I can see

34:21 What you’re saying, but to me, Mike, I think this is when we look at PowerBI and when you when you look like a dashboard, it looks so easy, but the work that gets behind it for a semantic model, the DAX, the planning, and honestly, that is going to be part of AI. Now, it could get easier, but honestly, the planning and that conversation piece of it, I think, is part of the skill and part of the planning. That’s part of the work. I don’t think we’re going to get to the

34:52 Point, maybe we do, but I’m not expecting or planning on getting to a point where I’m just going to give a couple of instructions, a couple back and forth, and we’re good. I think that’s a skill, and I think that’s also just the work that you do with AI. It is still work. And it is still a process because unless you have your own custom AI model that is already has the context for what you’re doing it’s going to require that especially when it’s something custom like let’s say you were building out something for Intelle right and you

35:26 Wanted to build out something where people can write or use a pen make it touchcreen I don’t know whatever it is right well unless it has all the context of the coding already there your style in that background ground. It’s going to always require that back and forth and until we actually have something more either you locally build it or it just you give it the proper instructions, we’re always going to be in that back and forth. look at C-pilot Studio. The amount of effort and and I’m not saying that in a negative way, but the amount of build and setup

35:59 To do a proper co-pilot agent is extensive because it needs to be. And I don’t think you’re going to I see what you’re saying and I want to get there too, but I think this is where the skill part of the the instructions and writing back and understand what it’s saying is so vital to this age of AI right now. I guess what I’m saying, let me let me give you another lens of how I’m thinking about this too. Okay, I watch a lot of YouTube videos around other individuals creating really rich, very smart custom visuals. They’re

36:32 They’re they’re PowerBI visuals, but they’re they’re highly stylized, right? In order for you to get a visual with dynamic text or maybe dynamic colors or dynamic fonts on the visual, you have to be able to put the visual on the page, you have to be able to say, “What columns do you want to have?” And then I have to build potentially a measure or two to help supplement that visuals coloring or styling or whatever. And so when I when I look at these visuals, I look at what I want to produce, right? I want the output of what they have on the video. What I don’t want to go through or what I don’t

37:06 Want to learn honestly is I don’t want have to learn the 16 or 15 clicks to get to that visual and how it looks. Let’s be real, right? [laughter] But that’s just laziness. No, it’s not laziness. No, it’s not lazy. I’m being I’m I’m being making a joke. But yeah, I know what you mean. You want I don’t want to learn that. I want I want the output of the visual, but I don’t want all the clicks. So where where is the AI to come in and say okay I want this visual with this style on it and it should know based on like these other visuals how to make that visual look what what I want like this is

37:39 Things I can do. It seems to me like when I look at like if we if we step back and said let’s look at all the reports that everyone has ever built across all of Microsoft there’s probably enough patterns to build any visual for everyone. Every every like all the everything that could be created has already been created in someone’s report somewhere. So this means all that knowledge is is captured in like a few heads that are at the top figuring this stuff out building. So to me, I’m looking at this

38:11 Going, well, where’s the agent to do that? Like I think sometimes we we th focus too much on like AI to like build the entire report. We need AI to build like the single visual, the small like let’s let’s go down supplemental a supplemental stuff like I should be able to describe more of what I want around that particular thing and it should be able to do all these 15 16 settings, clicks, whatever. It should be able to produce that visual very quickly. And we’re at a point right now where AI can build measures and things. Yeah. So, I’m where I’m at right now is I need to train myself to be better at doing

38:44 These things. , I’m going to continue to learn in this area. I think this is a very open-ended area. And as it relates to like building reports, as it relates to building pipelines, as it relates to building notebooks, there’s going to be patterns that are going to evolve. And I think 2026 is where I’m going to see a lot more of these patterns emerging with AI. And those are the things I want to stay close to in order to be more effective as a developer. I like this and I like this because I think this is we call it the year of the supplemental agent basically because yeah we can always build from the ground

39:16 Up but do what would make my life easier is those like to your said that add-on that marginal difference if I can do that quickly rather than having to build everything or provide that context from the ground up. Okay, I like that. I’m on board. I’m on board. So we’ll see what comes out. And a lot of this is also too you’re at the you are at the mercy of right now the AI companies to help build that too because there’s only so much we can do. I don’t know if that’s necessarily true yet. I don’t maybe it’s not that. Maybe it’s a custom model that has to be

39:50 Built. Maybe it’s they have to make that they have to release it. Well, why can’t I why can’t I funnel like why can’t I just collect all the videos of people doing this and just feed those videos to the AI and say, “Hey, look, here’s 30 videos that I have found of people doing small quick tutorials around how to build this visual. Why can’t the AI just look at that and figure out, oh, that’s what the video says to do. Here I am in desktop. I’m just going to click the buttons for you and do it.” Like, there feels like there’s potentially a gap here that or something along those lines. Like, so I I don’t know yet, Tommy. This is where I’m like, I’m not quite sure how it’s

40:22 Going to fit yet. I think you’re going to be building supplemental agent. , that sounds I like that. I like that. All right. So, let’s go on to my second and I Dude, I wrote a few down. So, where am I going to go here? what? I’m going to say it. I am going to enter, Brad, he sold me. And I was not a believer before, but Brad from Microsoft sold me. data warehouse is going to be part of my routine, man. , and that’s going to be something I’m going to be pushing. I think warehouses and what we learned and

40:57 What the amount of focus more importantly Microsoft is putting into warehouses. It goes back to what we say, see where the money is and see where they’re writing. And Microsoft is always, always providing major updates to the data warehouse, especially right now they’re nothing fancy. It doesn’t glitter. It’s but it’s it’s it’s the admin stuff but you need that you need that and so I don’t think it it gets to know people like AI does who oh AI now will actually say hello to you cool that’s so cool but not really helpful but all the things that

41:30 Brad has done and his team around and everyone else who’s working on it is it may not be glittery but man is it integral for how data flows and processes also what the data warehouse can do outside of a lakehouse you again I’m always going to be in that point, Mike, on what does this do better than that? Why would I do both of them just because of one has different features? It has to accomplish this better than X for me to use it. And I’m seeing more with the warehouse. I’m going to be diving in more to not just building

42:04 Them, but doing them the right way and making more of my data not just in a endpoint in in Lakehouse, but in a warehouse. Yeah, I’ve already started doing a lot more work around warehouse. I’m finding a lot of value from it. , I think I’ve also tried to start shifting some of my mindset a little bit too as well, Tommy. I I initially I was always using fabric as like the destination for transactional systems. I think I I think in 2026 there’s actually some opportunity here to make fabric the transactional system either like SQL databases, Cosmos DB. I

42:38 Think there’s also some other databases that are coming that are coming to fabric in 2026 that are going to make it a bit more of that as again as a developer I build apps I work with my team to build apps right and what we do is we stand up infrastructure that is behind Azure and we’re building servers and things that are over there as I’m looking at this I’m going well why do I need to build those things those those operational systems solely in Azure and then try to get an afterthought of bringing them over into fabric and I’m starting to really believe that small to mediumsiz applications might

43:10 Actually start needing to live like their data that sources for their application might actually live in fabric. How would that work? , Python and Spark is doing this pay-per-use pricing. It’s called autoscaling which really is more along the lines of pay for what you use. I think if you see more pay for what you use type features on like Cosmos DB, SQL Server, I think that actually removes the barrier to entry for a lot of companies to say, look, I’m not worried about bumping up to the threshold in fabric around, hey, I’ve outrun my capacity. I need to go

43:45 Buy a larger capacity to run my operational system because that’s a big no no. Like my one of my biggest hesitations about what you’re speaking to Tommy is I can’t use these systems for real applications if I’m ever going to get throttled. It just needs to continue to work and me as the developer need to make sure that we’re using it as efficiently as possible. So I can’t have Cosmos DB or SQL Server throttling queries to my actual applications if they’re if they’re going to be having problems. So having something that’s like

44:18 Autoscale on everything in fabric is going to be very essential for transactional systems to really move them into fabric. Does that make sense what I’m saying? No, I this is a this is 100% also for me too because I’m doing a lot more not necessarily from the application building but most of the organizations we work with they have applications that they’re integrating or talking to. So, it’s that importance there on understanding how that works and making sure to your point that we could trust it that it works all the time and that you’re not going to have, oh, it’s a demo or we’re still in a beta and it’s

44:51 Going to work. it’s either we use it or we don’t. It’s either going to work in all these situations or we have an alternative and that’s where our organizations are and we’re seeing that, we’re getting there and that’s a huge part for me too. But it’s also just the capabilities of the warehouse and it’s just another great skill set to have. you can get by with the lake, you can more than get b by with the lakehouse. And I I make sure I want to emphasize that you can do most of the daily tasks with the lakehouse and be great. That being said, that lakehouse is that underbelly that’s

45:26 The it’s the cave that can accomplish so much. So that is my second resolution. Mike, do you have a third? Yeah, I’m going to just react a bit more to your your warehouse comments. I think I think the warehouse thing to me I’m I’m maybe not as much warehouse related, but I I think Tommy to your point there, I’m really exploring now what more can I do with fabric in like operational systems that live on on top of fabric, right? So you’re saying warehouse specifically is an area of focus for you. An area focus in a similar way, Tommy. I’m I echoing the same thing you are, right? I’m also

45:59 Saying I still also want to learn more about warehouse things as well. But maybe mine’s a bit broader than that is like look I’m trying there’s the terms online analytical processing that’s usually the the term for the data warehouse and then there’s online trans and then online transactional processing y so l OAP and OLTP I think fabric is becoming more it the story is not just data and AI platform as like data analytics and I think Alex you’re pointing this out properly right it’s not just the analytics platform it’s analytics and transactional system

46:32 All merging together as one and now you can build things that are really close together and I think this is the same if I look at the story here and again this is maybe intentional maybe not intentional but this is how me as an outsider looking at Microsoft and what they’re doing is perceiving this when I would look at the early days of power query remember Tommy we’re doing power query with Excel and and then PowerBI comes out and we have this I literally have training for it this month yeah so okay good I whenever I do training on micros Microsoft and PowerBI. I I keep emphasizing Power Query is so useful to

47:06 The to the daily workflow of of a daily business analyst. Like if you’re doing any analyst work inside Excel, you better be learning Power Query. It’s so good. There was this moment in time I I specifically remember walking into a customer who was heavily heavily Excel. Everything everything was Excel. Everything everything in the whole company went from data systems, SQL servers, access databases, everything boiled down to an Excel sheet. So they loved it. And I remember talking to them and saying, “Look, when you have IT professionals building out data

47:41 Workflows, they’re doing transforms, they’re moving data around, they’re shaping things, and you’re in the business working in Excel, transforms, data manipulation, doing things. both teams from two different areas can now read and understand the same language. Power Query was this homogeneous language, this homogeneous tool that would allow the business to go in and pull tables from SQL, shape them, conform them, adjust them, and in a graphical way could give that back to the IT team and say, “Look, I realize you aren’t allowing me to have access to

48:15 The full data set set or system. You’re just giving me like a window, but here’s what I’ve did. Here’s my work. I’ve I’ve solved the math problem and shown my work. Here’s the work that I produced. Can you make another database or table or view that mimics the same length the sim the same transformations? And so to me, having those two worlds come together over top of the same tool was incredibly powerful. And for that company, everyone was like the light bulbs were clicking on. It’s like, oh yeah, this is an this is going to be a

48:48 Game changer for us. And so I see fabric being the same thing now again for both this like transactional side of the world and this analytical side of the world where we’re saying, hey, there’s these two worlds that have lived separately for so long, right? The tooling now is providing us a common platform. Tommy, I may not be playing in data warehouse as much. I may not be doing Cosmos DB. I may not be doing SQL database as a business analyst. That may not be my forte, right? But at least now I’m in the same ecosystem of tools or software or

49:23 Solutions where I can now directly connect what I’m doing to reporting. And so the the teams the two separate teams are now able to work in the same environment. And I think that removes a lot of potential data silos that we had previously. Let me emphasize your point, Mike, because some people may be listening and saying, “Oh, Excel and Power Query. Here we go again. Maybe he’s just pandering to Alex guy.” I literally have two monthlong trainings this week on Excel and Power Query. And here’s the best part. They just asked

49:57 For Excel and I had two scoping calls and I said, “We’re going to do Power Query.” This is because they were telling me, “What do you need like what are you guys doing in Excel?” They said, “Well, we’re trying to transform data. We’re trying to report on it.” I was like, “Here we go.” I said, I want to teach you like we’re going to focus on Power Query and Pivot tables and then we’re going to get into PowerBI and it’s still happening today. This is not something because we’re dinosaurs and we’ve been doing this for 10 years. We have, but it’s still companies are still relying on Excel for their data today. So, we’re this is still a large gap here or a wide range of organizations on

50:32 Where they’re at in their data maturity. So, you’re spot on. You’re not talking about something 10 years ago. you’re talking about something I’m gonna be doing on Monday [laughter] like exactly right and and I’m gonna I’m gonna also lean into your point too about where fabric is because this is also my last resolution and I’m going to dive into Mike that intelligence developer idea we talked about this on episode 487 I believe and this idea Mike of I am an int I’m a business intelligence developer or architect or whatever you want to call it but at the end of the

51:05 Day Mike like my role and what I should be focusing on is going to be data governance is going to be trust is going to be on the workflow it’s it is has expanded now the difference when I look at where I see fabric today compared to when it first came out I thought Mike that I was a data engineer and now I’m expanding the roles and I have to learn all these tools no my role is how does the data flow through an organization and is going to be that conductor’s hat more. But a big part of this too is still the trust

51:40 Because it’s how the data flows and you emphasize this a few times already today. Yep. AI is such a big part of that and it’s whether it’s through an application that they’re building or an data agent. And the only way you’re going to trust an AI is if you trust the data. The only way you trust the data is you actually have good maturity. So that’s going to be such an important part of what we do that we talk about the tools all the time. I hope we do a good job on the podcast of also emphasizing that gray area, the stuff you’re not going to find in documentation. It’s the people and the process side of this. But everything

52:15 We’re talking about to me, everything that you said just emphasizes more the people part of this. So that’s going to be a big part for me as we lean and we lean out here as we get into the new year. Awesome. It’s never been more important. I agree with that. I got one more quick personal one for those listeners out there. Sure. You’re gonna get a better Tommy. So, we hear your comments and Mike, I’m gonna I’m gonna open up here. I’m going to be personal with the listeners. We hear you people. Sometimes people are like, “Is he alive on the podcast? He looks a little

52:46 He’s rambling too much. What’s going on? I can’t understand.” Well, yeah. So, we hear we hear you. We hear you. I hear you. I definitely hear you. So, you’re going to get a better Tommy. I know. I’m sure that’s turned some people off, but no, this is going to be a good year for us. And I want to say first. Sorry about that. My bad. , [laughter] I’m sure that’s not great. But no, we I think for you and I, Mike, on the podcast, I think I want to lean in more on the intelligence developer and we we keep doing what we’re doing. We want to hear from you guys, too. We want to hear what do as we get into the new year,

53:19 What you guys are going to be getting into. We love the relationship in the community here, but no, you’re going to be getting a more involved, if you want it, a more involved Tommy. So, awesome. I do agree. , so Tommy, you’ve done you’ve done your three things, right? So, is it mind to Okay, so I know we’re getting close on time here. , the last one I think maybe I’ll point out here for news new year’s resolutions is an area of space that is relatively untapped is this whole new area of workloads. I’m building them right now. My company is spending a lot

53:50 Of time and effort around building more workloads and figuring out what’s effective there. , I really enjoy workloads and I think it opens up a ton of possibilities for companies and people to develop in there. I really like it. I think it’s it’s there was if you think about where external companies can invest inside PowerBI and in fabric. One of those areas was custom visuals. I don’t think Microsoft landed the plane very well in custom visuals. They didn’t do a good job of communicating licensing. It’s difficult for developers to to build on

54:23 Top of the custom visual space. It was very it’s also very difficult and I honestly this is just me personally. If you’re building a custom visual company and you’re charging $10 per custom visual per user, this is insane in my opinion. That’s that’s totally the wrong way to price these things. So, I think Microsoft tried to do a good job there, but these companies tried. It’s just overpriced. I’m not going to double my price for PowerBI just to use some custom visuals or even licensings. I think it’s totally crazy where I look at this and go where Microsoft has done a much better job is

54:56 Licensing and availability of letting us as developers use the fabric ecosystem APIs, users, user permissions. I’m a much better developer on top of the workload experience. I can much more easily integrate my product directly into the fabric ecosystem. So that’s something that I’m really looking forward to is just more development around workloads. So, one of my 2026 resolutions is spending more time and effort around building workloads quickly, getting them out into the marketplace, seeing what people like. , you’re going to see

55:28 More things come from PowerBI tips and and my company developing more workloads that I think are going to be effective. And we’re going to just throw a lot of stuff out there and see what sticks. , some of it will work, some of it won’t, some of it will keep enhancing, some of it we might we might not. We might not just leave it in preview for a while. So, , we’ll see what happens. We’ll take a a key here from Microsoft, but I do think workloads are very powerful and now with AI and Agentic things, you should be able to build those easier and faster moving forward. I can’t wait to comment on your workloads as well. So, honestly though,

56:01 Mike, if you do get to the point, I will create a page on the on the web that has your what’s in preview and what’s public. So, [laughter] no, I I like that. I think yeah it’s it’s you don’t really have much of an excuse if you had an idea for a workload at this point to not get at least try it with an if you especially have VS Code or GitHub Copilot to get started to get something out there. So I I I really like that. So that’s where we’re at right on on that front of things. With that being said, those are all my

56:33 Ideas. Tommy, do you have any final ones for New Year’s resolutions? No, I I think I said my pieces, so I think I am good to go. Okay, awesome. Well, that being said, I hope you have found some interesting things around what we’re going to do here. Stay tuned. Tommy and I are going to be ramping up hopefully more educational pieces here in the near the new year 2026. we want to continue to invest what we’ve learned. Tommy, I was just thinking about this while we’re talking about the podcast. You said we’ve been doing this for 10 years. So 2015, it’s now actually now it’s 11 years. We we just rolled over now. We

57:04 10 plus. You say 10 plus 10 plus years. So, between the two of us, you’ve got 20 years of experience listening to this podcast at your disposal here. , that being said, if I think about Tommy, think about your age 11 years ago, Mike, I’m looking in the video and I see the white hairs here. So, , my age 25 years ago. No, no, 10. Just just 10. [laughter] Just not 25 years, but just thinking 10 years back. We’ve been doing this for 10 years. I didn’t have kids, right? Isn’t that wild? Okay, so I just

57:37 Want to say we’ve been doing this for quite a while. It’s very been fun, very much fun for us to do this with you, the community. we do welcome another year with you. Very happy to continue on with the podcast. I think it’s I think it’s fair, Tommy. I think we can lay claim to this. I think we are the world’s largest Fabric and PowerBI podcast by number of episodes, by hours of content created. I think we’re the world’s largest data podcast for Fabric and PowerBI at this point. So, thank you all for a really wonderful 2025 and looking forward to a really fun and exciting 2026. U, we are going to continue to have guests on. ,

58:13 And Alex is already asking about when he can get back on the episodes again. So, yes, Alex, you are 100% invited back. We need to get you in for some more topics. , we’d love to have more guests on here. So, anyone who’s from Microsoft listening, if you want to get on the podcast, make sure you reach out to us. Go hit up our Microsoft forms over at PowerBI tipsodcast. You can check connect with us there. If you already know who I am and how to get a hold of me via Teams, you’re more than welcome to reach out to me on Teams as well. Tommy, where else can you find the podcast? Dude, you can find us everywhere. You can find us on Apple and Spotify and wherever get your podcast. A lot of my

58:47 Podcasts are going to Netflix. We haven’t been asked by Netflix yet, so maybe one day. But make sure as for now still to subscribe and leave a rating. It helps us out a ton. Do you have a question, idea, or topic that you want us to talk about in a future episode? Head over to powerbi.tipsodcast and leave your name so we can mention you and a great question so we can do it. And finally, join us live every Tuesday and Thursday, 7:30 a.m. Central, and join the conversation on all of PowerB. Tips social media channels. Thank you so much for a great 2025.

59:19 Looking forward to a really fun 2026. We’ll see you more in the future. Thank you all and we’ll see you next time. [music]

Thank You

Happy New Year and welcome to 2026! What are your Fabric resolutions? We’d love to hear what you’re committing to this year.

Want to catch us live? Join every Tuesday and Thursday at 7:30 AM Central on YouTube and LinkedIn.

Got a question? Head to powerbi.tips/empodcast and submit your topic ideas.

Listen on Spotify, Apple Podcasts, or wherever you get your podcasts.