Latest / Microsoft Cloud IT Pro Podcast / Episode 438: Taking your WorkOS and LLM Wiki to the Next Level
Transcript
- 0:00Welcome to Episode 438 of the Microsoft Cloud IT Pro Podcast, recorded live on October 2, 2026.
- 0:11This is a show about Microsoft 365 and Azure from the perspective of IT pros and end users, where we discuss a topic, more recent news, and how it relates to you.
- 0:21Back in Episode 429, Scott and Ben introduced LLM wikis. Today, they're circling back around to see what they've turned into after using them daily for a few months.
- 0:32Scott walks through the skills that now run his daily life, from a morning prioritization pass and meeting prep to meeting capture and end-to-end daily reviews.
- 0:42We also talk about growing a workflow from an on-demand skill to a scheduled skill, loops, and deterministic scripts.
- 0:51We also discuss different approaches Scott and Ben have to creating custom dashboards and command centers and how they interact with their data.
- 1:01We're live with another episode of the Microsoft Cloud IT Pro Podcast, live from our houses.
- 1:07Live from New York. It's Saturday night.
- 1:09Oh, no, that's a different one.
- 1:11Different show. We're not quite that popular, Scott.
- 1:13Someday.
- 1:14I know we look that good, and we sound that good, and we could clearly do a primetime, not really primetime, late night,
- 1:20NBC show. Better be careful. We're going to end up talking about copyrighted stuff.
- 1:24We could go see if we could end up on The Onion and the new InfoWars channel or something.
- 1:28We could just spin yarn and conspiracy theories about Azure and Microsoft 365.
- 1:34365. We could. We could come up with all kinds of conspiracy theories.
- 1:38My daughter is actually enjoying that right now.
- 1:40Do you know it'll be fine?
- 1:40Yes, my daughter's actually taking a class right now that's all about conspiracy theories, where they spend like a week, because she's homeschooled.
- 1:50So all the kids in the class go home from their co-op, spend a week like looking at different conspiracy theories, and then come back to class the next week and debate the various conspiracy theories.
- 2:01So it's kind of interesting. She's having fun. They did the moon landing today. So the last week, she's been looking at all the different conspiracy theories around the moon landing.
- 2:09My oldest son is, he's studying history in college. And so he's, so he's in his second year now when he's getting kind of like more into things. And he has had the realization that history repeats itself.
- 2:27You didn't learn this in like middle school or high school, but you're coming to it in college. Okay, well, good realization.
- 2:32I mean, at least he came to that realization.
- 2:35There's some maturity there. I have hope that he'll move out of my house someday. We'll see.
- 2:40So you know what else happened today, Scott? I called my HVAC company because we have like in Florida, AC is kind of important. So twice a year, we have like regular, we have a plan. We sign up for discounted maintenance and all that. But then they do like twice annual, twice annual maintenance checks. And I got a text message and an email and all of it. Hey, it's time to schedule one.
- 3:05So I called them today because usually that works better than trying to do it online and all of that. And it was like, you're on hold. And then I had someone picked up the phone and it was like, I can't remember the name. Hi, I'm Lisa. This is your AI assistant. Like it was an AI.
- 3:23Well, he should announce itself that way. Person. I'd rather fall into like a phone tree and interactive voice recording like that, where it's very upfront and clear that yes.
- 3:35Yeah. And that's what it did. Name. But it was a little, it was interesting. I wanted to be annoyed at it the whole time because I knew I was talking to AI, but it actually went really smooth.
- 3:46They superimposed normal office sounds in it. So I could hear like a keyboard clicking in the background and all the stuff. But this AI assistant was, why are you calling? I was like annual maintenance and asked me all the questions it needed.
- 4:00Like had an actual conversation with this AI agent, just like I would have had with anybody in their office and was able to schedule my appointment, got the text message and the email with the confirmation of the appointment date and time.
- 4:14And as much as I wanted to be annoyed knowing I was talking to AI, it actually was a good experience. And I was pleasantly surprised and impressed with how well and smoothly it went.
- 4:28Good to hear. When you were texting me about it and telling me about it, I thought you were going to tell me like you fired off because there's lots of personal agents out there right now.
- 4:36So you've got Meadow with Muse, there's some invite stuff like Instinct that's out there. OpenAI announced Dots at their dev day this week.
- 4:45So I thought what you were going to tell me was like, oh, you had your AI personal assistant, because now these things are starting to come with phone numbers, and do inbound outbound calling.
- 4:55You had your AI assistant call another AI assistant, and maybe they got caught in one of those loops of like talking to each other in like binary or like whatever.
- 5:05No, but that would be funny.
- 5:06Sound and wavelengths. They wanted to have like a waveform representation to go and talk back and forth.
- 5:12I should try that next time is have my bot call their bot.
- 5:16Speaking of bots and Dots, did you also see the somewhat trolling, hilarious news like OpenAI announced their Dots?
- 5:25And immediately after that, Musk went out and bought Dot.com and redirected it to the GrokBot.
- 5:34If you want a domain name about a product, you're going to buy the domain name before you announce it.
- 5:38Or you just get big enough that you have your own company domain name, right?
- 5:43Like the whole like Cloud, Dot, Microsoft, and things like that.
- 5:46See, that would solve it too.
- 5:47That's the other way to go about it.
- 5:49But yeah, we should talk about AI stuff today.
- 5:52We should.
- 5:53I want to revisit the talk that we had a little while ago about getting started with LL and Wikis and kind of talking about what they can turn into with a little bit of maturity.
- 6:10So like this will be a little bit more of like a freeform conversation for us.
- 6:15I want to talk about my setup.
- 6:17I want to learn a little bit more about your setup.
- 6:20And I really want to dig into kind of the workflows and maybe like the skills and agents that you're using.
- 6:28Like what are the harnesses that you're using around this that make it helpful for you and empower your day-to-day work, accelerate your day-to-day work.
- 6:39Maybe they make things harder, right?
- 6:41As you've been experimenting and going through.
- 6:43So what are things that maybe you hit a brick wall with or I hit a brick wall with?
- 6:47And we kind of went in a little bit of a different direction.
- 6:52So yeah, you want me to go first?
- 6:54Sure.
- 6:55What I've been doing?
- 6:55Yes.
- 6:56I'm curious to hear what you've been doing.
- 6:58So I don't want to get to your point earlier about like copyright infringement, things like that.
- 7:02I don't want anything there.
- 7:03But I took what were like various instantiations of LL and Wikis that I've been working with around like various areas of like research for specific projects and initiatives that I work on.
- 7:17Just like my day-to-day tasks, tracking, things like that.
- 7:20And I consolidated them all into one wiki and it's my WorkOS, right?
- 7:27It's my work operating system.
- 7:30So it's meant to track really everything that I do across work, whether that's working.
- 7:37So I'm a product manager, so that could be working on a new product.
- 7:41It could be working on an update to an existing feature.
- 7:44It could be like building a new feature.
- 7:46Maybe that's not a full product, right?
- 7:47But it's complimentary to existing product suite, things like that.
- 7:51And I do all the PME things that come along with that of market research, writing PRDs, specs, having customer conversations, wrangling work management systems like ADO.
- 8:08We use Azure DevOps.
- 8:10Maybe somebody's using JIRA, something like that.
- 8:13But really, everything that I do gets logged, gets tagged, and goes into that WorkOS.
- 8:24So off the cuff, that sounds like, okay, but how do I get everything in there?
- 8:29And what's important and what happens along the way?
- 8:33So I've got a set of skills that are out there that I've put together that work for my way of work and that operate within the harness that I use.
- 8:44So the harness that I use today is Microsoft Scout, the desktop app for Scout.
- 8:50I'm going to be a little bit sad when that goes to autopilot here in the future because I don't know what that's going to do.
- 8:55Not autopilot because we have Microsoft autopilot.
- 8:58We need to differentiate now, Scott, because we have Microsoft autopilot.
- 9:01Autopilot inside of copilot.
- 9:03Copilot autopilot.
- 9:03And copilot autopilot.
- 9:05That's another topic for another day.
- 9:07We'll get to that one eventually.
- 9:09So I've got this set of skills that help me with inputs and outputs to get information into that wiki and then help me get good concrete outputs out of it.
- 9:23So one would be, for example, one of the ones I use quite a bit actually is an email thread ingester skill.
- 9:34So my work takes place across teams, email, in-person conversations, teams meetings.
- 9:41If I'm meeting with customers, it could be Zoom meetings, things like that.
- 9:45But particularly for internal conversations, a lot of that stuff still happens in email or there's newsletters and various things that go out.
- 9:53So I have a little skill that's my WorkOS email thread ingester.
- 9:59And because it works in context of WorkIQ and the Microsoft Graph, it respects sensitivity labels, DLP, all those kinds of things.
- 10:09It's not like I'm taking super secret stuff out of the system and throwing it over in this random text wiki over here that lives in my OneDrive or anything like that.
- 10:18So I use that one quite a bit.
- 10:19And that's super handy and helpful for me for point-in-time kinds of things.
- 10:25I have a skill that runs every single morning for me that goes every single weekday for me.
- 10:32It runs every morning at 9 a.m. local, which is my WorkOS daily prioritization skill.
- 10:42So what that does is it looks at the to-dos that are in my WorkOS.
- 10:48So I made this very conscientious decision not to use Planner or Microsoft to-do or anything like that.
- 10:55I'm just kind of managing everything in markdown files inside my WorkOS.
- 10:59So your source of truth for your tasks is these markdown files in WorkOS?
- 11:04Yes.
- 11:05Okay.
- 11:05But I find that works because I can tag them.
- 11:08They carry the tags through the wiki.
- 11:10They carry all the grounding and the context.
- 11:14I can link to-dos very easily to projects or to initiatives and various other contexts that exist in that wiki.
- 11:22So what my kind of daily prioritization skill does for me is it's meant to look at my prior day and or my incomplete tasks that are sitting out there that I've already spec'd out and help me kind of rationalize,
- 11:38here's the number one thing that you should do today out of your existing to-do list based on what happened between yesterday and today through your email, through your team signals, all those kinds of things, documents people shared with you.
- 11:52Here's another task that you should go and do.
- 11:55And then if it has room for it, the way I structured the skill is I always want to have kind of like three things on my plate in front of me that are like these are like the must-dos for today kind of thing.
- 12:05So that sits out there.
- 12:06I have a question about that one because I have one similar to that.
- 12:09Have you, and this is something I've done with mine with some success, have you had that skill when it looks at your daily to-dos, also look at your meetings for the upcoming day to prioritize if there's anything that you have to have done for a meeting coming up in like the next eight hours based on your tasks and projects as well?
- 12:30Yes and no.
- 12:30So we should get into that one next.
- 12:33All right.
- 12:33I want to kind of start the conversation here talking about the skills that I've integrated and that I'm using within Scout as a harness.
- 12:40I could be using Copilot, CLI, Claude, Codex, Anti-Gravity.
- 12:45It really doesn't matter, right?
- 12:46A skill is a skill and that markdown is portable and can come over and I can tie my connectors, all that kind of stuff.
- 12:52But there's this concept that I think becomes very powerful of moving from skills to loops and having loops that go out and execute things for me.
- 13:02So while like my daily prioritization skill is just a skill, I could run it on demand.
- 13:07I could come in and I could say, hey, run this thing.
- 13:10It's both a skill and it's a loop for me.
- 13:13So I put it in the harness and I schedule it and I say, hey, run this every day at 9 a.m.
- 13:18I actually have, I can run the same skill like every 30 minutes basically on a constant loop and have it do constant reevaluation for me like throughout the day and pivot things through.
- 13:30So I have two other skills when it comes to meetings that also function as skills and or more harnessed loops and kind of just on the flywheel of they're going to do things automatically for me.
- 13:44So one is a meeting prep skill.
- 13:48So I have like WorkOS-MeetingPrep and I have WorkOS-MeetingCapture.
- 13:54So WorkOS-MeetingPrep is my prep skill.
- 13:57So as my daily prioritization skill is running in the morning and it's going through like that harness and that automated scheduling and now it's turning more into a loop, what that's actually doing, one of the subtasks that it has is it goes and runs meeting prep for every single meeting that I have that day.
- 14:16That's on my calendar, okay?
- 14:18And then as I'm coming out of meetings, many of the meetings we attend are recorded, they have transcripts, they have all of those things.
- 14:27I have a meeting capture skill.
- 14:29And my meeting capture skill goes and evaluates the recording, it evaluates the transcript, it grabs the WorkIQ context, which is inclusive of like the meeting chat.
- 14:38So you can catch like documents that were shared in the meeting, things like that.
- 14:42And that comes through and effectively it turns my meetings into a two-step flow of prep and then capture.
- 14:50Or if I have a meeting that maybe, let's say like I was an optional attendee or maybe it was a learning forum, it's something that I want to catch up on later.
- 14:58So I wasn't active in it.
- 15:00For like meetings on my calendar that I'm just following or I'm an optional attendee, things like that, my morning prioritization skill doesn't actually run prep for those automatically.
- 15:09It just skips over them, right?
- 15:10I don't need them right now.
- 15:11So for those, if I want to, I can go in and I can basically run, I kind of have this like overarching maybe superset skill, which is generate and capture.
- 15:22So I can run generate individually, I can run capture post-meeting, or I can run generate and capture, which kind of does both for me.
- 15:29And generating capture is helpful for the meeting that's already happened that I didn't attend because they do different things, right?
- 15:36Prep goes out and looks at my inbox, it looks at conversations with meeting attendees, and it says, hey, what's the context for this thing?
- 15:44Capture is go blend that kind of meeting brief and the meeting prep with the actual outcomes of that meeting and what's next along the way.
- 15:55And then I have another skill that runs in a loop, again, on an automation, which is my end-of-day wrap-up.
- 16:04And part of my end-of-day wrap-up is that skill evaluates my wiki and it says, what are the meetings I attended today that I forgot to capture?
- 16:12Or I forgot to do a prep and capture on, and it just goes and grabs them automatically for me and kind of brings that context in.
- 16:19And that's good for me because then in that virtuous cycle of tomorrow's a new day, my automation for my daily prep runs again.
- 16:27And now it's able to catch up on the meetings and the context and things that are there.
- 16:31So I don't have to constantly be like human intervention in the loop to say, like, click a button or go run this skill, go run this skill, go run this skill in my harness.
- 16:40It's just happening kind of automatically and automagically as I go through my day.
- 16:49And then, let's see, because this is an LLM wiki, like, if people are on, like, the Carpathi-style LLM wiki path, they often end up with, like, that raw or clippings or ingest folder for things.
- 17:00Well, I'm a PM.
- 17:01I do research.
- 17:02I have to keep up on industry news.
- 17:04It could be Gartner, IDC, launch from a competitor, things like that.
- 17:08So I'm just running things in a markdown wiki at Subsidion.
- 17:12So I can use Subsidion Web Clipper.
- 17:14So I send things into a, like, a raw or a clippings directory.
- 17:18So I have a, hey, go ingest my clippings.
- 17:21Like, go rationalize over this.
- 17:23And again, organize it in context of my projects.
- 17:26Tag it.
- 17:27Do all those things that you need to do along the way.
- 17:31And then I have these kind of, I don't run them in loops.
- 17:34I don't run them automatically.
- 17:35I have a bunch of, I call them, like, my little, like, sub-agents or my helper skills.
- 17:40So I'm responsible for a portion of a business.
- 17:44That business has, like, a monthly business review.
- 17:48We do ongoing monthly product reviews.
- 17:50We do weekly execution reviews, things like that.
- 17:54And those often require, like, very structured inputs of, let's go grab your ADO items, get the current status.
- 18:00What's the dates, what's the update, things like that.
- 18:04So I have these skills that go out and help me do those things, like WorkOS MBR prep, WorkOS product review prep, things like that.
- 18:13And they all go out and they kind of work and they help me get the rough outlines for those things.
- 18:21I don't let them, like, write everything for me.
- 18:23I have this very visceral reaction to, like, just pure, like, AI-written things.
- 18:28I'll let them bullet it out so that then I can take them and kind of put my own thinking or my own spin on top of it along the way.
- 18:38And then the last thing that I did, and this is probably, like, super niche, but it's been very effective for me,
- 18:47is I took the things that I was doing in the harness.
- 18:53So I started in Copilot CLI, eventually moved to Scout.
- 18:55Now Scout's kind of like sitting over there.
- 18:57So Scout's really like my automation engine.
- 18:59It's not actually where I go and run these things from.
- 19:02So if I need to go, say, run a meeting prep, so I've got an upcoming meeting and that prep hasn't been run yet,
- 19:09or I've got to go run a capture, I built myself a WorkOS dashboard.
- 19:14So I have a little local HTML site that's served to me, and that site presents to me my calendar.
- 19:24It presents the meetings that haven't been prepped versus those that have been prepped,
- 19:29those that haven't been captured versus those that haven't been captured.
- 19:32It gets my to-dos there for me.
- 19:34So rather than me having to go into Obsidian, say I'm updating a to-do from not started to in progress,
- 19:41I actually, I go to this website, and I just click the to-do.
- 19:44It's kind of like to-doist or something for me just sitting there.
- 19:47And I update it right there, and that thing goes and writes the wiki, the markdown for me.
- 19:51So I don't have to go, like, find the markdown file and worry about, like,
- 19:54oh, did I get the tag right, everything.
- 19:56Things are very deterministic for me.
- 19:59So I've got the wiki there when I need it, when I need to go search for it.
- 20:03I do all that through Obsidian.
- 20:04I've got Scout sitting on the side, which is running my loops and running my automations
- 20:09and is kind of the house for those skills in the execution engine.
- 20:14And then I've got this just kind of dashboard that I live in.
- 20:18And from the dashboard, I can do things, like, I've got, like, my calendar in there.
- 20:20Like, if I went and looked at my agenda view today, it would have this podcast recording
- 20:24as a meeting on there.
- 20:25I can click on podcast recording, and then I've got a little button there that says,
- 20:29open your brief, and it'll just pop me open straight into Obsidian.
- 20:32Like, it takes me right into the vault, right to that thing, and gets me ready to go and
- 20:39where I need to be.
- 20:41So that's been super helpful and super contextual for me.
- 20:44Because when I come in the morning, I refresh my homepage for my WorkOS, right?
- 20:49I just hit the refresh button.
- 20:52And then all of a sudden, I'm presented with, here's your daily priorities.
- 20:55Here's your top three to-dos for the day.
- 20:57Here's your agenda for the day.
- 20:59Here's how these priorities relate to your agenda for the day.
- 21:02Here's all the things that I generated for you.
- 21:04Here's how many aren't left.
- 21:06Things like that.
- 21:07It's a very freeing thing.
- 21:10Like, in the past, I would have sat there and said, like, oh, crap, I've got a meeting.
- 21:13Let me go to that meeting.
- 21:14And you kind of come in all discombobulated, and you do the context switch, and you come
- 21:18out to the other side.
- 21:18Then you go, okay, I've got to go to my next meeting.
- 21:20I don't have time to go back and kind of write my notes up or capture them, anything
- 21:24like that.
- 21:25Now it's all just turnkey.
- 21:27It's all right there for me and ready to go.
- 21:33And also, very deterministic.
- 21:36So the last time that we chatted about this, I believe I had also mentioned kind of taking
- 21:42skills that maybe start out with a bunch of overarching shape in them, but then you want to make them
- 21:48more deterministic over time.
- 21:50So sometimes you have to stop relying on the model and the skill itself.
- 21:55And you need to kind of write scripts and have a script do the generation for you so that
- 22:00things are always formatted consistently, they come in the right way, all that.
- 22:04So for me, my meeting prep skill, it started as a very generic skill.
- 22:08Like, go find the transcript and the recording for this meeting on my calendar and capture it.
- 22:13And then I refined it over time, but it was still a bunch of markdown.
- 22:16I actually took it and worked and iterated on it back and forth a little bit, and I was able
- 22:21to get it.
- 22:22So now my meeting prep skill is really a script that calls WorkIQ.
- 22:28And it calls WorkIQ in a way that the output is very deterministic for me for every single
- 22:33meeting I go to.
- 22:34It's got the meeting attendees, it's got the transcript overview and the summary, it's got
- 22:39here's the key items from the chat, here's the documents that were shared, things like that.
- 22:44And what that allowed me to do by putting that skill, and now the skill is just a wrapper for
- 22:50calling that script, back to that dashboard thing, right?
- 22:53That dashboard's there, like I can't click a button on that dashboard and have it go run
- 22:57a skill.
- 22:58But I can have it go run a script, right?
- 23:00And kick that off very easily, like it just goes and runs a Python script in the background.
- 23:04So whether I'm running the skill through the harness, through Scout or through GitHub Copilot
- 23:09CLI, or I'm clicking the button, at the end of the day, the skill and me clicking the button
- 23:14are calling the same Python script, giving me the same deterministic output every single
- 23:20time.
- 23:20It's interesting to hear you talk because I've done some things very similar, but then
- 23:24I've also done some different things.
- 23:26Like I said, it's a very empowering thing, I think.
- 23:28So this is in no way, shape, or form a badge of honor.
- 23:32I'm actually kind of like scared when I say stuff like this out loud.
- 23:36I have on average 65 meetings a week on my calendar.
- 23:40That's nuts.
- 23:40So that's my average.
- 23:42Some weeks I have 80, some weeks I have 40, but I'm sitting right around 65 on average
- 23:49over the past three months since I started tracking this.
- 23:52I do not attend 65 meetings a week, to be very clear.
- 23:55Like 20 of those are probably just like learning or like FYI things.
- 23:59But having that skill out there, that prep and capture skill, like that still allows me to
- 24:04ingest the knowledge and have it out there.
- 24:05Like we had a training last week on like GitHub Copilot CLI and like sub-agent usage and things
- 24:12in it.
- 24:12I was like, oh, that's cool.
- 24:13Like I really want to go catch up with that.
- 24:15In the past, I would have completely forgotten about that, right?
- 24:17I would have had it written down on a piece of paper.
- 24:19I would have had like a to-do and to-do to go watch the recording for that.
- 24:23And I never would have gotten to it or come back around.
- 24:25But because I have this like prep and capture thing, it actually can sit there and contextualize
- 24:30for me and not only get me the information I need, so I can go get the summary and decide
- 24:34like, oh, was that actually worth my time or not?
- 24:36But then it's got all the canonical links back to the full recap and everything that I need
- 24:43out there and that's ready to go.
- 24:45So I'm probably like on some further end of the pendulum swinging.
- 24:50Like I haven't quite found equilibrium there where I'm probably still capturing too much
- 24:54stuff and I'm bringing in too much context.
- 24:56But I'm still at the point even like a couple months into this where I'd rather have more
- 25:00than less.
- 25:01And I hope I can get to like the less is more kind of thing along the way.
- 25:05But when it's so easy to do, because the loops are there and the harness is there, I'm kind
- 25:10of inclined just to keep it running.
- 25:12And then, like I said, it makes a bunch of this stuff like more powerful for me, right?
- 25:16Because if I'm going and like we're working on maybe like a new like CLI component or we're
- 25:23doing an MCP thing or some kind of like agentic thing and now I've got all these training on
- 25:29how agents work and how they do stuff, like that all applies, right?
- 25:32That all comes back into what I need to do.
- 25:36So contextually, that stuff could be part of my research for a SPAC or a PRD like that
- 25:43something later.
- 25:45And then I have these aspirations to go build some more loops.
- 25:49And I very much want to do this.
- 25:50So for me, for example, one of the things that I like to do is I like to go on Reddit and
- 25:58LinkedIn to go find feedback from users.
- 26:02So I've got a couple different places.
- 26:04Azure Storage really doesn't show up on Hacker News for some reason.
- 26:07They must not think we're a startup or something.
- 26:08I don't know.
- 26:09But we get like decent feedback on forums like Microsoft Q&A and Reddit and things like that.
- 26:15So I would love to go and build like a user research loop that can help me go like research
- 26:20my product on like a daily or weekly or monthly basis, things like that might be a little
- 26:24bit harder for me.
- 26:25I saw just today.
- 26:26So today is October 2nd.
- 26:28I think I saw earlier that Reddit is going to shut down all their external search access
- 26:33and all their RSS feed access.
- 26:35So that's going to make it a lot harder for agents to get at the data that's in Reddit.
- 26:38So maybe it's a good thing.
- 26:40I haven't built that loop and skill yet to go and do that.
- 26:44But the whole thing of starting with skills, automating those skills, and then figuring out
- 26:50how to compound them into these loops that just kind of keep running until you get to the
- 26:54outcome you want, super powerful.
- 26:57If you're managing file storage today, you've probably felt the pressure.
- 27:03Refresh cycles are harder to plan.
- 27:05Costs and complexity are increasing, especially as file workloads scale across regions.
- 27:10That's why Microsoft customers trust Nasuni.
- 27:13As an Azure certified software, Nasuni provides a global file system designed for enterprise scale
- 27:20environments, keeping file data secure, governed, and accessible across Azure environments without
- 27:25added infrastructure complexity.
- 27:27It's the foundation to keep your unstructured data stable, even as your infrastructure evolves.
- 27:33To see how much you could save while keeping your initiatives on track, visit nasuni.com
- 27:39I have a few that I run regularly.
- 27:50I have not gotten too far into the loops yet.
- 27:54I have a lot of stuff that I still just run manually.
- 27:57I have a few things that I run as routines or scheduled jobs.
- 28:04But it's been, yeah, it's a lot of the way I would look at these LLM wikis.
- 28:09And even again, it's probably taking it further than what the LLM wiki did, like with your dashboard
- 28:14and HTML.
- 28:15I actually built my own custom Obsidian plugin that I have as a dashboard.
- 28:20So my dashboard sits right in Obsidian rather than a web page.
- 28:23How did you do that?
- 28:24I'm intrigued.
- 28:25I told Claude, so I actually told Claude to go build me a custom Obsidian plugin.
- 28:31And Claude has built the...
- 28:32Okay, built it as a plugin.
- 28:34So it renders in the canvas.
- 28:35It built it as a plugin.
- 28:37So it renders as a canvas, like down the left side of Obsidian where you have your different
- 28:41icons.
- 28:42I now have an icon for my dashboard.
- 28:45And my dashboard has a bunch of cards on it.
- 28:47So this is where similar type of concept with mine, only I combined a bunch of stuff.
- 28:53I have multiple LLM wikis in the same vault because I have one for work.
- 28:57I have one for some personal stuff.
- 28:59I have a separate podcast one in my vault.
- 29:02Because sometimes for me, the overlap between a topic we talk about on a podcast and a blog
- 29:10post, maybe I want to write for my website and a client that I'm working with, there's so
- 29:15much overlap there.
- 29:16I want them to be separate to an extent, but I also want to be able to correlate certain
- 29:22things across them.
- 29:23But then my dashboard has cards that are color-coded and labeled.
- 29:26So I have intelligent cards for my top to-dos, maybe things that are on fire, CRM type information,
- 29:35potential jobs, where jobs are in the pipeline.
- 29:38I have some podcast stuff where I pull in podcast stats so I can see trends with the podcast
- 29:45downloads and episodes month to month.
- 29:49I have some personal stuff where I actually pull in stock market.
- 29:52I have a markdown file where I put stock tickers I want to track, and then I have a dashboard
- 29:58that every time I go to it, it'll go out and pull the latest stock prices and put it on
- 30:03the dashboard.
- 30:04But then I put little buttons on it that'll call a skill to go do analysis on that stock
- 30:10ticker.
- 30:10So I have my own analysis on stock tickers within my LLM wiki.
- 30:16And I might have to explore this because one of the things...
- 30:18It's interesting.
- 30:19I am definitely juggling between Scout, which sits out there.
- 30:23It's a desktop app, right?
- 30:24It has to be running.
- 30:25So I've got Scout sitting out there running and...
- 30:28Wow, that hurt.
- 30:30Sorry about that.
- 30:30It was quite a whine by that dog, by that puppy.
- 30:33I've got Scout out there running.
- 30:35I have this website that kind of runs.
- 30:37It's just a pin tab in Edge for me.
- 30:39It's always my first tab sitting at the top.
- 30:42And then I have Obsidian.
- 30:45So having just one less thing, like I don't have to go to the website and I can just live
- 30:49in Obsidian.
- 30:50Because I do this thing where I've been tempted to...
- 30:53Maybe I was thinking about it wrong.
- 30:54I was actually tempted to see if I could find a way to embed a Markdown viewer in an Obsidian
- 30:59view in my website.
- 31:00But I should just totally go the other way.
- 31:02The Canvas thing is very interesting.
- 31:04So I'm glad you brought that up.
- 31:06I've seen a bunch of people doing similar things recently.
- 31:10So if you're in VS Code and you're using, say, like GitHub Copilot chat in VS Code.
- 31:16So VS Code has plugins, all those kinds of things.
- 31:19You can absolutely use your LLM to help you develop a plugin that draws on the canvas in
- 31:24VS Code.
- 31:25The other one is GitHub Copilot.
- 31:27So GitHub Copilot has a CLI.
- 31:29And there's a new GitHub Copilot app as well.
- 31:33And that app renders canvases in HTML as well.
- 31:37So I've seen some folks in my organization that are creating custom canvas apps for GitHub
- 31:43Copilot, the desktop app, because that's where they're living.
- 31:46And that's where they're doing their work out of kind of thing.
- 31:49So that's another super powerful concept is like kind of coaxing an LLM and getting it
- 31:54to build the view that you want and the experience you want in the place that you work.
- 32:00So going back to your skills, I have a lot of similar skills.
- 32:03Like I have a daily review skill that will go back and look at all my meeting notes from
- 32:08the day.
- 32:09It'll look at my calendar.
- 32:11Mine actually goes into, I don't know if you've tried this, it'll go into my team's
- 32:15conversations.
- 32:16And I've told it to look for certain, there's certain teams that I'm in that are super chatty
- 32:22where only like five or 10% of the stuff is relevant to me.
- 32:26And I'm like, go look through all of these chats and teams and give me, tell me where there's
- 32:32a chat related to a certain topic because that's the only thing I care about.
- 32:36So part of my review will actually analyze all my team's chats for me and tell me if there's
- 32:41anything I should go back and look at my team's chats or if I can just ignore all of them.
- 32:45I run a similar one.
- 32:46That's another automation or loop that I have out there running.
- 32:50So it looks across my inbox and my team's chats.
- 32:53And your team's.
- 32:54And it looks for emails from my manager, my skip, or people on my team.
- 33:00Okay.
- 33:00Who report to me.
- 33:02And initially that was a, let's get out there and kind of grab everything, right?
- 33:10Like the signal to noise ratio was, it was really bad.
- 33:13It was pretty horrible because like anytime somebody chatted with me on my team, like they
- 33:18could ping me on the side and say, how was the weather today?
- 33:21And I would all of a sudden get this alert that popped up and said, Joe is, Joe is, I
- 33:26don't have Joe on my team, but Joe is asking you how the weather is today.
- 33:30Do you want to respond?
- 33:31Oh no, that's, that's not important, right?
- 33:33So I definitely had to go and refine that thing in a similar way.
- 33:36I have a little bit of contextual filter, things like that.
- 33:39So it only alerts me now when folks are kind of pinging me with urgency or high priority
- 33:45items, particularly things that align to work streams that were kind of
- 33:51integrated on or that I need to know about what's happening.
- 33:54I've actually been thinking, and one of the things I would love to do is take it to the
- 33:56next level and just have it be able to respond back on teams.
- 34:00So like if somebody says, oh, hey, can you give me the latest update on feature ABC?
- 34:08If I already sent out the status update for feature ABC this week, I just wanted to send
- 34:13them the doc or send them a link like, or send them like the subject of that email that went
- 34:19out or whatever that was.
- 34:20And that feels like low hanging fruit and something that I could go out there and automate fullness
- 34:26of time.
- 34:27I just haven't been sure if it's worth it yet.
- 34:29Like putting the effort in to have an additional contextual filter, definitely worth it.
- 34:33Like I said, like I was getting all sorts of like stuff, like very much along the lines
- 34:37of like, so-and-so is telling you the weather's nice today.
- 34:41Like very important.
- 34:42Please respond.
- 34:43Like, no, I don't need to do that.
- 34:44I don't find mine too.
- 34:45It's just certain topics.
- 34:46But then in my dashboard, so I can run all of my skills from my dashboard.
- 34:51I can just click a button in the dashboard and it'll go open a bash prompt, a Claude session
- 34:58in the background and go run the skill in the background and I won't ever see it.
- 35:03So any skills that I don't have to interact with, then I have other skills where I know
- 35:07like I may go back and forth with them on or have more of a conversation with Claude.
- 35:13Those, the button actually copies a prompt to my clipboard.
- 35:16So I just open up a Claude session and paste the prompt right into a Claude session.
- 35:21Interesting that you chose to go down that path.
- 35:22Everything out of my dashboard.
- 35:24Go open the harness, go open Claude and run the skill versus me where I was like, I'm going
- 35:30to go, I'm going to go have it write the script and have it be super deterministic so I can
- 35:34run that script from a skill or from a button or from here.
- 35:38Some of my skills do still run scripts.
- 35:41It's the same.
- 35:42I still have a lot of scripts where those skills, it's still deterministic.
- 35:46It's a button, open a Claude session, run a script, run a skill.
- 35:49So one of the things maybe I was going at and something for folks to think about is, and
- 35:53one of the reasons I kind of go down this path quite a bit, is optimizing for token usage
- 35:58and context.
- 36:01So if you're using Claude as a harness, Claude has a fairly large system context already coming
- 36:07with it as part of the harness.
- 36:08And every time you're invoking a skill, you have to load that context no matter what.
- 36:13Like that's there.
- 36:14And so sure, you should have the advantages of cash tokens and things like that.
- 36:19If Anthropik's doing their job right, if they're not doing their job right on a given day, or
- 36:24stuff does go sideways, you're really in this place where you're burning tokens.
- 36:29And this stuff does cost money, right?
- 36:33Like, I'd be curious if you're comfortable saying like, do you have a sense for how much
- 36:37yours costs to run over the course of a month?
- 36:40If not?
- 36:41I do other things in it.
- 36:42And I mean, I will say I am paying for Claude team now.
- 36:47So I'm paying for the team edition of Claude.
- 36:51What's the team edition run?
- 36:52Which it's somewhere between, so they have like the Claude max, where it's the 5x or the
- 36:5820x.
- 36:58I'm actually sitting at like 6.5x.
- 37:02They don't have a 5x or 20x.
- 37:05So I would say I'm equivalent to the 5x, but it's a little bit more.
- 37:08I'm paying $125 a month for Claude.
- 37:11So my token consumption input and output is 100 to 120.
- 37:16I don't have the hard average in front of me, but I would say I probably average.
- 37:20I probably average like $110, $115 a month, something like that, to go through and run
- 37:26everything that I need to do.
- 37:28But that includes like the things like, hey, going and like bootstrapping the dashboard and
- 37:33things like that.
- 37:33I would expect like continued optimization as things come down.
- 37:38I was trying to find this the other day.
- 37:39Over the last 30 days, I can see things like product stickiness.
- 37:44This is connectors.
- 37:45I can see the skills.
- 37:47Most of these run in Claude code, usage, top skills, tool usage.
- 37:53I was just trying to look super quick if I could see how many gives me value.
- 38:00I haven't found.
- 38:02I'd have to dig through here like how many tokens I use.
- 38:07It's not free.
- 38:08No, it's not free.
- 38:09But I can fit it all in there with room to spare because I run these skills, but I also
- 38:13use it all the time for writing PowerShell scripts or analyzing data, that type of stuff.
- 38:18Because I do, similar to you, I have a lot of these skills that have turned into Python
- 38:25scripts in the background as well, where the skill calls a script to do a lot of the work
- 38:31for it.
- 38:32Python seems to be where it all lands.
- 38:34But a lot of what I do is very similar type of stuff that I built to you, except that
- 38:38I have multiple LLM wikis in that vault.
- 38:41And then I did the dashboard in Obsidian.
- 38:44It's been, at least from my end, super empowering because I get to work the way I want to work.
- 38:49I'm always frustrated at the friction of, here's Microsoft to do, and it does to do the way
- 38:55Microsoft wants to do them for that product or things like that.
- 39:00So I would love it if I could say, I could take this thing and package it up and give it
- 39:06to you and it would work for the way you work.
- 39:08I don't think that's the case, nor do I think it's the point.
- 39:11The whole point is to have the way you work and the set of skills and loops and sub-agents
- 39:18or authentic flows and connectors and everything else that are out there to wire everything
- 39:25up and get it going and have a good set of outcomes for it.
- 39:28I really do feel like it helps accelerate the work that I do.
- 39:33I think it keeps me better grounded.
- 39:36I still have to go through a lot of context switching and I find it super helpful to have
- 39:39a lot of the grounded notes and things just right there at my fingertips.
- 39:43Go ask a question so I can search in Obsidian or when I do need to use an LLM to iterate
- 39:49over it, great, here you go.
- 39:51I can hop in that directory and invoke my harness of choice and I'm off to the races.
- 39:58I did things a little differently too for my tasks because I have contractors working for
- 40:02me.
- 40:03So I want to make sure tasks are updated in Asana and some of that.
- 40:08So my dashboard, instead of managing tasks here, all of my skills and prompts and everything
- 40:14ensure that Asana stays the source of truth.
- 40:18So if I see something on my list of my top to do or something that I need to get done, it
- 40:23actually links to Asana so I can click it, open it in Asana and I check it off there and
- 40:29then the dashboard will update.
- 40:31And same thing for on fire.
- 40:33I have two right now where there's an email and a Teams conversation.
- 40:37I have created my dashboard so I can click on those items that I've pulled in and it takes
- 40:42me right to the email or it takes me right to the Teams conversation.
- 40:46I have one for blog posts that takes me right to the draft blog posts or to the client record
- 40:53in my CRM.
- 40:54So my dashboard, the way I've thought about mine, I've turned my dashboard and my LLM into
- 41:00an integration point so I can pull all the information from all the systems I want and built a central,
- 41:07dashboard command center that I can work from but then link out to all the different
- 41:12things that I need to respond to that pop up on it.
- 41:15I'm on a very similar path.
- 41:17I have this kind of backlog of things that I want to go kind of integrate next.
- 41:22So one is, I mentioned we use Azure DevOps, ADO, for kind of tracking of epics and features
- 41:28and PBIs and things like that.
- 41:30I'd love to have that integrated view right there for, hey, I'm working on project XYZ,
- 41:36click on project XYZ, here's the ADO items for it, and here's their current state and status,
- 41:42things like that.
- 41:44The other one that I want to build is, at some point, I struggle with this stuff, I don't
- 41:50know when I'm reinventing the wheel.
- 41:52This one feels a little bit maybe like reinventing SharePoint, but I've got these projects that
- 41:56exist.
- 41:57So when you do the Carpathia-style LLM wikis, you end up with artifacts and entities, you end
- 42:02up with these discrete templates for types of objects in your LLM wiki.
- 42:08So I've got projects and work streams and events and things like that.
- 42:13So for my work streams and my projects, there's these kind of canonical briefs that sit there
- 42:18that have all the detail about them.
- 42:21And then they have all the wiki links to sublinks and things like that.
- 42:25But sometimes the sublinks they carry are links to documents, so like Word docs, Excels,
- 42:30PowerPoints.
- 42:31So I kind of want to view where I can say, hey, for Workstream ABC, or Workstream A, click
- 42:39on Workstream A, here's all the important documents for it.
- 42:42Like, here's all the contextual things and just like the little snippet and summary of
- 42:46it that's there.
- 42:47And then like the document, to be very clear, like the artifacts, the canonical artifact,
- 42:50not stored in my wiki.
- 42:52Like that stuff's still IRM, it's got sensitivity labels, all that stuff.
- 42:56But the link to it is in my wiki.
- 42:58So now it's like one-stop shopping for me, where in the past, I would have been going
- 43:02to SharePoint and just going to kind of like the generic intranet search or even like Copilot
- 43:08search.
- 43:09I'm finding it to be kind of annoying because it's an LLM-based search.
- 43:12So you kind of start talking to it like an LLM and it goes sideways pretty quick.
- 43:17So I've got all this stuff sitting.
- 43:19Like the context is there.
- 43:20It's just about, like you said, integrating the views and kind of making it one-stop shopping
- 43:25across all these various systems.
- 43:27Yeah, my project for my dashboard has a backlog markdown file in it where I just, when I come
- 43:32up with an idea, I throw an idea in there.
- 43:35So when I have some time or I'm sitting on the couch in the evening and I want to build
- 43:38a new feature on my dashboard, I have my whole backlog and I can pick one off the backlog
- 43:42to start building.
- 43:43Anything we miss.
- 43:44I'm sure this is one we're going to come back to because I feel like, to your point,
- 43:47you're always evolving it, changing it, modifying it as everything changes.
- 43:53And as you come up with ideas.
- 43:54The builder's itch comes in along the way.
- 43:58And then the other thing I like about it is, so you not only get to go build these things,
- 44:04but like I said, you get to adapt them to the way you work.
- 44:07And you also get to adapt them to your taste.
- 44:12Like I said, I don't have the greatest reaction sometimes when I have to go into a Microsoft
- 44:19to-do or a planner, but it doesn't rub me nearly as much the wrong way going into this
- 44:24thing that I built for the way I guided an LM to build it for me, because there's no way
- 44:29I was writing all that stuff myself to go and put it together.
- 44:32So it actually, it's like, it's a little bit more calming, right?
- 44:36Like while it's bringing in probably, like I said, more stuff.
- 44:40And if anything, like it's allowing me to over-function, but it's at least allowing me to over-function
- 44:46in like my way of over-functioning without like my head constantly on a swivel.
- 44:52That's the biggest advantage I've found of this and frustrations I've had other times
- 44:56that I'm like, how do I integrate Asana with Teams, with Outlook, in a way that works for
- 45:01how I want to work?
- 45:03This has allowed me to pull that information and put some integrations in place that A,
- 45:09I can never develop or build on my own.
- 45:12And B, I can use products, but then tie them all together in a way that makes sense for
- 45:18the way I want to see information, work with information, manage my day.
- 45:22Et cetera.
- 45:23So that's the part I've enjoyed about this, to your point.
- 45:27It's, I can tweak it to me.
- 45:29So with that, just wrap it up.
- 45:31We went almost an hour today, Scott.
- 45:33Sometimes we go a little bit longer.
- 45:34So I'm always interested in feedback from other folks.
- 45:37We've mentioned a couple of times and it's good to see people coming and commenting on
- 45:41Reddit or LinkedIn.
- 45:43I would love to hear what other folks are doing there.
- 45:46Like, what's your superpower, your unlock for LLM wikis?
- 45:53Like, are you using them?
- 45:54I think you and I are maybe a little bit more on like one end of the spectrum where we're
- 45:59like in CLI harnesses or a little bit more advanced thing.
- 46:04Like, I've seen folks on definitely like YouTube using very much like Claude Cowork style kinds
- 46:11of things to go and execute these same workflows.
- 46:15So I'd love to understand where others are experimenting, what they're doing, where they're finding the
- 46:20unlock, and maybe what else we're missing that could be like the take it up another notch.
- 46:25We'll do the final tap thing.
- 46:27Like, does this go to 11?
- 46:28Yes, yes, it does.
- 46:29Yes.
- 46:30Let us know.
- 46:31We'd love to hear it.
- 46:31All right.
- 46:32Well, that's Scott.
- 46:33It's time for the weekend.
- 46:34Almost time for the weekend.
- 46:36But yeah.
- 46:37Well, as always, thanks, Ben.
- 46:39Not a problem.
- 46:40Thank you.
- 46:40And we'll talk to you again soon.
- 46:43If you enjoyed the podcast, go leave us a five star rating in iTunes.
- 46:47It helps to get the word out so more IT pros can learn about Office 365 and Azure.
- 46:53If you have any questions you want us to address on the show or feedback about the show,
- 46:57feel free to reach out via our website, Twitter, or Facebook.
- 47:01Thanks again for listening and have a great day.