# Why Full AI Automation Fails in Production | Angela McNea

Markdown copy of https://realalpha.app/v/DwrYJCnoHZs/ on Real Alpha, the best minutes from long interviews with the people building AI. Cite and link the page, not this copy.

Angela McNeal argues that high-stakes AI needs controlled autonomy, explicit decision logic and human intervention—not full automation.

- Show: MTS
- Hosts: [Theo Jaffee](https://realalpha.app/people/theo-jaffee/) & [Sophia Dew](https://realalpha.app/people/sophia-dew/)
- Guest: [Angela McNeal](https://realalpha.app/builders/angela-mcneal/) is CEO and cofounder of Thread AI.
- Published: Nov 8, 2026
- Length: 29 min

## The best minutes

- [7:13](https://realalpha.app/v/DwrYJCnoHZs/?t=433) [Model routing depends on explicit business logic](https://realalpha.app/s/model-routing-depends-on-explicit-business-logic/) (3 min)
- [25:27](https://realalpha.app/v/DwrYJCnoHZs/?t=1527) [AI workflows improve through iteration, not instant automation](https://realalpha.app/s/ai-workflows-improve-through-iteration-not-instant/) (2 min)

## Also worth watching

- [4:25](https://realalpha.app/v/DwrYJCnoHZs/?t=265) [Enterprise AI stacks can fail as providers and teams change](https://realalpha.app/s/enterprise-ai-stacks-can-fail-as-providers-and-teams-change/) (3 min)
- [13:22](https://realalpha.app/v/DwrYJCnoHZs/?t=802) [Clinical AI needs a deliberate choice about which errors matter](https://realalpha.app/s/clinical-ai-needs-a-deliberate-choice-about-which-errors/) (4 min)
- [10:13](https://realalpha.app/v/DwrYJCnoHZs/?t=613) [Thread's Army DevCom research targets manual deconfliction](https://realalpha.app/s/threads-army-devcom-research-targets-manual-deconfliction/) (3 min)

Skip the Lovable ad, the roughly one-minute mid-show sponsor block, and the closing sponsor reads.

## Full analysis: Angela McNeal says broad AI decisions struggle beyond constraints

- In a hypothetical reconciliation example, mistakes can cost 'like, billions of dollars,' McNeal said.
- Her account puts workflow policy ahead of routing mechanics, but offers no comparative deployment results.
- Start with constrained tasks, record human exchanges, and measure what improves before widening scope.

Its 6 sections, what it means for you and what to watch are on [the page](https://realalpha.app/v/DwrYJCnoHZs/#analysis), free with an account:

- [12:17](https://realalpha.app/v/DwrYJCnoHZs/?t=737) Reconciliation illustrates the downside, not a measured loss
- [8:26](https://realalpha.app/v/DwrYJCnoHZs/?t=506) Business rules define what a router should optimize
- [5:44](https://realalpha.app/v/DwrYJCnoHZs/?t=344) Production strain appears after teams assemble the stack
- [14:48](https://realalpha.app/v/DwrYJCnoHZs/?t=888) Clinical eligibility turns metrics into a policy choice
- [26:38](https://realalpha.app/v/DwrYJCnoHZs/?t=1598) McNeal expects constrained tasks to improve iteratively
- [1:30](https://realalpha.app/v/DwrYJCnoHZs/?t=90) Thread sells control infrastructure, not outcome evidence

## Summary

McNeal is CEO and cofounder of Thread AI, which she says builds infrastructure for agents and workflows in high-impact settings. She previously worked at Palantir and writes the Handshake Protocol about how people and AI systems work together.

McNeal argues that broad, unconstrained AI automation struggles in production, especially when a wrong decision carries financial, legal or physical risk. She says Thread AI provides infrastructure for controlled autonomy, with explicit decision gates, human intervention and records of entire transactions. She sees workflow design—deciding what matters, which trade-offs to accept and when people should step in—as harder than building the underlying model router or software blocks. McNeal says teams should start with bounded workflows, capture human-AI exchanges and improve systems iteratively rather than expect full automation from the outset.

### Minute by minute

- [0:49](https://realalpha.app/v/DwrYJCnoHZs/?t=49) **Thread builds controlled autonomy.** McNeal says Thread AI provides infrastructure for agents and workflows in settings where a wrong action or system failure can have serious consequences. Its platform lets organizations limit what systems can do, add human intervention points and record the whole transaction, rather than govern only individual actions.
- [3:21](https://realalpha.app/v/DwrYJCnoHZs/?t=201) **Production systems need operability.** McNeal says she moved from an FDE role at Palantir to product work, while cofounder Maya worked as a software engineer on the product side. She says companies were buying separate tools for specific business functions or assembling their own systems, which could be hard to operate as providers changed, teams turned over or workloads grew.
- [7:13](https://realalpha.app/v/DwrYJCnoHZs/?t=433) **Routing depends on decision logic.** McNeal says the difficult part of model routing is defining the business logic: which models to choose and what to optimize. Teams must specify what counts as a hard problem or an expensive one, including whether the cost is money, time or people’s time. Thread lets customers build routers and decision gates into its platform.
- [10:13](https://realalpha.app/v/DwrYJCnoHZs/?t=613) **Defense work raises the stakes.** McNeal says Thread was designed as a dual-use company and describes research with Army DevCom on manual deconfliction, helping scientists and warfighters get accurate information at the right time. She contrasts this with financial reconciliation, where mistakes can cost a business billions of dollars and create legal risk, and says defense errors can be life or death.
- [13:22](https://realalpha.app/v/DwrYJCnoHZs/?t=802) **People design the trade-offs.** McNeal says engineers at large institutions are shifting toward managing, reviewing and designing systems rather than implementing them. She uses a clinical-trial model to show how people must weigh the risk of admitting a bad-fit candidate against the risk of excluding someone who could qualify. She says AI does not remove the legal, regulatory and human factors that shaped existing processes.
- [18:35](https://realalpha.app/v/DwrYJCnoHZs/?t=1115) **Workflow design is the hard part.** McNeal says Thread’s building blocks modernize established distributed-systems security and software practices rather than discard them. She sees workflow design—such as deciding who receives triage alerts and how much alert volume is useful—as harder than assembling the blocks. Thread’s platform is self-service, with an applied team that works with customers; she says spending on consultants to assemble AI stacks points to a gap a platform can fill.
- [20:57](https://realalpha.app/v/DwrYJCnoHZs/?t=1257) **Enterprise APIs need infrastructure.** McNeal says Meta’s Toolformer paper showed language models using simple APIs, but enterprise systems involve thousands of software providers, APIs, parameters, protocols and authentication schemes. She says the idea of flexible decision-making in software inspired Thread’s vision for horizontal infrastructure. Her problem-solving approach combines scientific method and hypothesis testing with empathy for customer experts, lessons she connects to her upbringing, studies and Palantir experience.
- [25:27](https://realalpha.app/v/DwrYJCnoHZs/?t=1527) **Full automation fails in open-ended work.** McNeal says decision-making is difficult to automate across a whole process unless the possible paths are tightly constrained. She argues that AI workflows should start iteratively: use building blocks, capture the human-AI exchange and build data the system can learn from. She cautions against expecting to automate every decision from the outset.

## Transcript

MTS, hosted by Theo Jaffee and Sophia Dew, with Angela McNeal.

Transcribed by machine (Whisper) from the episode’s audio, so a word may be off here and there.

[0:00](https://realalpha.app/v/DwrYJCnoHZs/?t=0) I believe in like intelligent systems and that we're building intelligent systems and using intelligent systems. But I think a lot of the decision making that I've seen attempted to be automated across the stack, across the process, be kind of unsuccessful, except when the feature space is so constrained. But when you like open up the aperture, I've seen things struggle still. Hello, welcome back. I'm here with Angela McNeal, who is the CEO and co-founder of Thread AI, which builds infrastructure for enterprise AI workflows. She also previously spent six years at Palantir and writes the Handshake Protocol about how people and AI systems work together.

[0:50](https://realalpha.app/v/DwrYJCnoHZs/?t=50) So welcome, Angela. Great to have you on. It's a pleasure to be here. Thanks for having me. Yeah, so we were talking about a few things before we went live, but I guess, first of all, wanted to hear about Thread AI and how you guys are thinking differently about making progress with AI. So at Thread, we provide the infrastructure that AI agents and AI workflows can run on when an action or a wrong action or a system failure or a bad decision can have critical impact. So financial risk, reputational risk, legal risk. So we built our platform on the thesis of controlled autonomy. So how do my agents and workflows reason and act in a system that you control, that you limit, where you can inject humans at any point in time of that process.

[1:48](https://realalpha.app/v/DwrYJCnoHZs/?t=108) Either the agents will reach out to them or you've injected in by policy and then everything is over, is captured in a record. So a lot of companies govern individual actions, we govern sort of the whole transaction. Sure. Okay, okay. So what are some of the, like, failure modes or some of the problems that you were seeing that you wanted to address? Yeah. So we saw a couple of different patterns of companies sort of rebuilding and reassembling stacks when they're thinking about putting AI into production. Maya and I talk about this a little bit about how we were, you know, my co-founder, we were inspired to start the company when we, you know, read this interesting paper. But when we saw, okay, like there's this bottlenecks we're shifting from building models and training models to, okay, you have models, like now what do you do with them? And when you look at an enterprise and when you look at the real world, you're like, okay, a real enterprise has, you know, tens, hundreds of different softwares. You have different identity providers. You have different kinds of data you have different kinds of risks and rules and regulations um and we saw people sort of again assembling and reassembling all these different building blocks and pieces they needed to take

[3:04](https://realalpha.app/v/DwrYJCnoHZs/?t=184) these models and put them in meaningful useful ways and wasting a lot of time um so we're like okay well how can we provide um sort of the building blocks that are necessary so that you don't have to rebuild isolation and metering and identity and logging and governance and you can and just focus on, you know, what you're trying to do. Right. So I'm curious, did your time previously at Palantir as an FDE, is that when you started thinking about, like, how do we put these things into production or, like, into real-world environments as opposed to just purely having the technology? Yeah, Maya and I joke a little bit.

[3:39](https://realalpha.app/v/DwrYJCnoHZs/?t=219) So my co-founder and our CTO, she and I were at Palantir together for four years. But, so I joined as an FDE and then switched to the product side after about a year, I spent the rest of my time there. Maya joined exclusively as a software engineer on the product side. And it's funny. We joked that we were spoiled by solving hard problems. Right, okay. Because of the types of things, the kinds of, again, problems that we were exposed to and that we were working with. When we started our company together, that sort of same ethos, those tenets of, okay, how do we provide the right tools to solve critical problems and power mission-critical workflows and agents?

[4:27](https://realalpha.app/v/DwrYJCnoHZs/?t=267) That was in our DNA. i see yeah that makes a lot of sense so do you find that there were like companies that were sort of developing their own like um in-house systems and they were just sort of like reinventing the wheel over and over or were there like b2b companies that were all kind of doing the same thing like what what did you see specifically we saw kind of two buckets of things of like okay companies okay i want to solve invoice processing i want to solve procurement automation i want to solve, right, like marketing automation, I want to build my reconciliation flows, whatever it may be, buying individual sort of vertical tools specifically for those things.

[5:08](https://realalpha.app/v/DwrYJCnoHZs/?t=308) So you're assembling and your vendor management problem goes pretty wild. And then we were seeing a pattern where companies were like, okay, cool, we've built some really neat things in those individual business units. Why can't we take that and why can't we move them somewhere else? Right. So that was sort of bucket one. And then sort of bucket two is, okay, everyone can be a builder now. And I don't, that's like not a pejorative, that's exciting. And the nature of software development is changing. But when, you know, everyone's a builder and people are creating a lot of these systems, you can build these big capability maps, but the operability of a system is often, right, a big challenge.

[5:50](https://realalpha.app/v/DwrYJCnoHZs/?t=350) So, how do you, when you construct and build these, like, sort of, like, assembled stacks with all these different pieces, these building blocks, what does it look like when, you know, you change your sort of, like, you know, coding provider behind the scenes or when this team turns over or we've, you know, all of a sudden we need to load test this dramatically and we were seeing a lot of those stacks sort of fall over. Okay, I see. So, what is the operability of the system that's created? Yeah, when did you start to see that this was a problem? I think this was like, you guys raised in 2025? We raised our A in 2025.

[6:30](https://realalpha.app/v/DwrYJCnoHZs/?t=390) So Maya and I left in spring of 2023. Okay, yeah, yeah. Early in AI days still. Yeah, our thesis from the beginning was like, the, you know, the glue, the harness, right? The control, the governance, all of these pieces were going to be important. And it's been fun to sort of like watch your hypotheses play out live. And, you know, we've always been sort of, you know, pragmatists in how we build. And we're like, okay, this is a challenge for people now at the best of times.

[7:03](https://realalpha.app/v/DwrYJCnoHZs/?t=423) **Angela McNeal:** And this is just going to be compounded when builder ecosystems grow, model providers expand and grow.

[7:10](https://realalpha.app/v/DwrYJCnoHZs/?t=430) You're going to have frontier models and sovereign models and local models. It's just going to compound. Very prescient bet that you guys were making. So I'm very curious about what inspired you to write Every Good Decision is an ensemble because we see now, we talk a lot about model routing and how it's very important to have something that can choose what level of model based off of some sort of cost trade-off. And then you zoom out and make the argument that everything has always been this sort of ensemble and not just, like, one model or entity making decisions. It's, like, whether it's, like, how the judicial system works or whatnot.

[7:50](https://realalpha.app/v/DwrYJCnoHZs/?t=470) So, yeah, so I'm curious, how does that connect to Thread AI? Yeah, so we were, again, I was seeing this explosion. conversation around routing, and there are real important conversations to be had, and we're seeing it with our enterprise customers of like, when is it too expensive to use this kind of model, or the latency is too high, or there are like real trade-offs around the

[8:16](https://realalpha.app/v/DwrYJCnoHZs/?t=496) **Angela McNeal:** specific nodes. And the interesting part to me is that like the actual hard part, the technology behind a router I don't think is as interesting as like the application the hard part

[8:32](https://realalpha.app/v/DwrYJCnoHZs/?t=512) is the application around it how the business person or the engineering teams decide what is the logic of who what models do we pick what like what are the things that we're trying to optimize and then you have to pre-define all of the decisions up front saying like okay well we've decided that this is what a hard problem is. This is the shape of a hard problem is. We've decided what is expensive? How do you define expensive? Is it expensive in time, in people's time, in money? So you have to predefine all of these business and esoteric and abstract things ahead of time before you even get to, cool, now we just need something to help me route between these systems. And so we built our platform so you can build these kinds of routers natively in our product, but we spent a lot of time on those decision gates explicitly on those logic gates because we were seeing like in the real world, those were the actual barriers to putting something in production. Yeah, I see. I want to shout out our sponsor, Lovable. You know, the app you've been meaning to build, the internal tool, side project, or product you'd otherwise lose a weekend too? With Lovable, you get software you can ship today. That includes an editable codebase

[9:48](https://realalpha.app/v/DwrYJCnoHZs/?t=588) you can inspect and change, plus two-way GitHub sync. Lovable also handles the backend and infrastructure, including managed Postgres, auth, storage, hosting, payments, and more, so you don't have to wire it all up yourself. And through Lovable's MCP server, you can create and deploy your project directly from the agents you already use. Turn ideas into software people love with Lovable. Lovable.dev. Now onto the episode. So I think like, you know, the topic of agent failure can be very high stakes, specifically with anything to do with defense or anything to do with national security. So I'm curious about your, you know, your like defense threat AI and how it helps in defense use cases or defense providers. Yeah. So being a dual use company was very important to me and Maya from the beginning. And there's, like, no more demanding customer than the, you know, U.S. Department of War, Department of Defense. And our same primitives, our same building blocks that we use in sort of those environments are the same, the power, right, or like our enterprise work. And I could talk about sort of in the beginning, Maya and I sought out hard problems we want.

[11:04](https://realalpha.app/v/DwrYJCnoHZs/?t=664) Our goal was to help companies and organizations and institutions, like, fix hard problems and help people. So a lot of our work, the one that we sort of talked about publicly recently is we are doing some research with, like, Army DevCom,

[11:23](https://realalpha.app/v/DwrYJCnoHZs/?t=683) **Angela McNeal:** and a lot of that work is around, like, manual deconfliction, right? So how do you help people get accurate information at the right time when you're sort of building and maintaining existing systems,

[11:36](https://realalpha.app/v/DwrYJCnoHZs/?t=696) control systems, and innovating new ones. So how do you help the scientists, and then how do you also help, like, the warfighters on the ground? Mm, yeah. And, again, that's probably one of the most demanding environments I can think of. Yeah, totally. I mean, I'm just curious, like, if you could paint a picture of, like, the stakes for agent failure in, like, this industry, but also others. Yeah. so So, again, an enterprise version of this is if you have agents that are helping you with your sort of reconciliation, again, a common one, reconciliation flows in sort of a big financial institution.

[12:17](https://realalpha.app/v/DwrYJCnoHZs/?t=737) Mistakes there can cost your business, like, billions of dollars. So they carry a huge, like, financial and potentially, probably, actually, like, like, a legal risk element to it. And then in a, again, in a sort of defense setting, it's like, you know, those are life or death stakes. Yeah, yeah, yeah. So you have extrema. Yeah, definitely. So I'm curious, too, like, what's next for Thread AI? So we're growing, sort of, quite a bit. We have a bunch of open roles around some of our public sector work in particular. So you'll be seeing more from us on that element. And then one of the things I'm pretty excited about is that sort of the investments that we've been making around our building blocks.

[13:07](https://realalpha.app/v/DwrYJCnoHZs/?t=787) So for companies who are selling products or building products that they're selling to their customers and want those products to be more adaptable and agentic and AI forward, like we've been providing sort of building blocks for companies building products. So keep an eye on some of the stuff we're putting out there. Yeah, awesome. So I'm curious, like, what, from, you know, looking at these companies, if you have any takes on sort of the ways that companies are just, like, structurally changing and maybe the role of, like, human agent interplay. Like, what, from your perspective, like, where are we going?

[13:46](https://realalpha.app/v/DwrYJCnoHZs/?t=826) Yeah, it's a good question because I couldn't say that there is one singular pattern. And we work across industries, so we work across financial services and manufacturing and hospitality, advertising. So there's an interesting blend where we're seeing the nature of the developer change. So at these large institutions where, you know, engineering is now moved towards more of a like, I'm managing and reviewing and designing as opposed to implementing. And we're seeing, I think, that kind of rough same pattern happen with like knowledge workers and sort of leaders and experts where the construction of the system, the construction of the process is often the intellectual work.

[14:34](https://realalpha.app/v/DwrYJCnoHZs/?t=874) So like we were talking about designing those gates, you as a business user have to define what is important. Is it, actually a good metaphor is something I used to tell my old team at Palantir. When you're thinking about evaluating a model, and let's say the model is like, hey, this is determining who can join this clinical trial. Okay. And when it's a classic sort of, it's a classification model, you're saying like cool like yes or no in each camp you can join this trial you can't join this trial when you're evaluating it um you have to think about the you are looking at metrics accuracy

[15:13](https://realalpha.app/v/DwrYJCnoHZs/?t=913) **Angela McNeal:** and precision and recall like you're saying hey like and then you as a business person with the scientists have to decide what's more important is it more important to bring some have the model

[15:27](https://realalpha.app/v/DwrYJCnoHZs/?t=927) be wrong and bring someone into this clinical trial and they get these drugs and they're not a good fit and it can damage their health or is it more important to you know not include someone miss out on someone and so then they you know get this thing and then sort of so like how we wait to those true deaths what's more important um that is often codified into the systems um so it's never like yes or no it's an investment someone makes in the logic as they're building these flows yeah um and so i think that's a sort of we're seeing more intentionality focus there of like what's important to us how do we want to spend our time and how we want our systems okay i see i see i i think that like my um my fear or like what would be an intuitive fear is maybe if you're at a corporation and you're um you know you have a lot of people who are now working on software yeah um and like and i guess that's like now become a flatter skill that everyone can you know partake in to some extent that there would be uh less intentionality behind And I guess it's different with industries that are very high stakes, but I'm curious, like, you're seeing that it's becoming more intentional with how people are designing sort of like the logic of things.

[16:38](https://realalpha.app/v/DwrYJCnoHZs/?t=998) On the, like, big picture process, I think as for an individual's knowledge work, I think that it's like we're still seeing, like, Wild West, right? Yeah, yeah, yeah. Like, the individual's productivity tools. But when you think of, like, okay, this is like an actual process, um often it's designed that way historically for a reason a combination of people legal regulatory like there are a lot of factors went into designing that thing from the beginning and so if you're looking to disrupt it and if you're looking to automate and inject ai at different components you still all of those factors don't go away you just now have to sort of rethink them in this like ai era so we're seeing again a lot of enterprises spend a good amount of their ip focused on that sort of redesign um which has been pretty cool to see yeah but i do think it is still again like wild west yeah we'll continue monitoring right after this message from our sponsors 11 labs ai that communicates at human level across every channel and modality 11 labs dot io slash mts scale your startup on neon millions of developers and startups have already chosen Neon for their backend. Start on the free plan or get up to $100,000 in credits for your

[17:52](https://realalpha.app/v/DwrYJCnoHZs/?t=1072) startup at neon.com slash MTS. Special thanks to our sponsor Kong, the AI connectivity platform. Connect APIs, LLMs, agents, and systems with serious security and governance. KongHQ.com. Finally prove what your AI investment is worth with Lairdyn. Head to Lairdyn.com today to start measuring your AI's bottom line. MongoDB, trusted to run critical applications by 75% of the Fortune 100, the intelligent data platform for the era. Apply today for access to up to $15,000 in Atlas credits, dedicated technical support, and exclusive events and go-to-market opportunities. MongoDB.com. Chargebee, billing and monetization infrastructure for the AI era. Chargebee.com.

[18:39](https://realalpha.app/v/DwrYJCnoHZs/?t=1119) I wonder when that will like the dust will begin to settle there yeah um yeah so I'm curious also is there anything that you find like what is the hardest to automate or even with the building blocks that you were mentioning like um what is difficult about designing building blocks you kind of have to create something that is that will work across many different like teams or whatnot I'd say that is mine and Maya's superpower in that, like, our primitives that we built sort of from the beginning have been able to adapt to all of the sort of technological changes in the past sort of four years.

[19:16](https://realalpha.app/v/DwrYJCnoHZs/?t=1156) Um, uh, and, um, we were very explicit from the beginning, um, to not throw away, right, like 20 some years, 50 years of distributed system security, um, right, software best practices, but to kind of like modernize and make them more adaptable. also learn from best practices. And I think the hard part is often the piece, again, that comes after of like, you know, you can build a triaging workflow, right? But then you as the builder of this workflow, how do you decide who gets these alerts, the volume of those alerts? Are they too noisy? Are they not as important? So the actual workflow design, I think, is the more...

[20:00](https://realalpha.app/v/DwrYJCnoHZs/?t=1200) challenging part. Yeah. Um, do you find that you can transfer these across industries as well? Um, we, so we built our platform to be self-service. Um, so our customers are a lot of our builders, but we do have like our applied arm too, we'll work with our customers and we'll share best practices and knowledge and sort of construct with them. Um, I was thinking about, I saw there was that right enterprise AI spend is raw allegedly rocketing or booming. Um, I think what we're seeing probably a big piece of that is there's this been huge investment we're seeing in like consulting services where people are spending a lot of money asking for help and having those people assemble and build those stacks for them which to me is a perfect indicator that there is like a gap for a platform to fill yeah and that but but people do you know want help Yeah. So tell us more about the platform and how you went about designing it. Did you design this from the get-go? Was the vision always, we're going to do a platform?

[21:02](https://realalpha.app/v/DwrYJCnoHZs/?t=1262) Yeah. We built horizontal infrastructure past life. Wanted to build it now. And it's funny. I think Maya and I talk about this, I think we've talked about this publicly, of the thing that really inspired us was this paper that came out of Meta. Sure. And it was the Toolformer paper where they were like, okay, we have a language model and we're going to train the last couple layers to speak simple API. And there it's like, I think one was like a calculator or like weather or whatever. And if you're a software engineer and you look at that, you're like, well, that's not how an enterprise, like enterprise APIs don't work that way, right?

[21:39](https://realalpha.app/v/DwrYJCnoHZs/?t=1299) There are thousands of software providers at an enterprise. Each of these providers might have thousands of APIs with hundreds of different parameters. You have different protocols and different authentication policies and schemes. You have different data types. The ecosystem is so complicated. But the idea that, like, cool, I could put flexible decision-making in software was pretty exciting. And so we're like, okay, how do we take that? And then you play it out N years. You're like, what could it look like best case if you have these again sort of intelligent systems in and across a company you're like well to do that and to make that a reality working backwards from that you're going to need a lot of infrastructure yeah and so that was the vision from the beginning yeah how do how easily do you find it you personally to like extrapolate all these trends when it comes to like enterprises and their use of technology i think that um if you are building in and as like as a product person if you are building in this space and leading a company in this space that is like table stakes that is a skill, a superpower that you have to have if you're selling also to horizontal so across these different patterns

[22:54](https://realalpha.app/v/DwrYJCnoHZs/?t=1374) and that also too is the fun part of it the pattern matching I get to see 10 different kinds of workflows a day across different companies and again the fun part is you have all this information how do you pattern match and how do you pull out the interesting pieces and right like fun product work yeah yeah um i'm also curious too because you seem to think about logic a lot and and so like what was your um we talked briefly too about like your background in math computer science but like what is the lens that you view problems in and maybe like to at palantir what was the lens that they view problems and how did that all tie together um i think you know i've i have i grew up You know, I'm the child of two climate scientists, grew up in a household where you're always deconstructing and talking about things at dinner tables.

[23:46](https://realalpha.app/v/DwrYJCnoHZs/?t=1426) So there's a lot of the scientific method being applied sort of throughout my life. And then spent, I was at Columbia where I studied computer science and math. And so there is an engineering element to science. There's a scientific element to how I approach problems. So how do you deconstruct problems? How do you determine what the base case is? And how do you hypothesis test? So there's that element to that. But I think there has to be, we don't live in a vacuum. We don't live in like a simulation with like a topical hot take. Yes. With, right, perfect actors applying perfect behavior, right? We live in a deeply messy and human world. And so the way that you approach problems, you have to approach them with empathy and understanding and try to like solve the problem. That's where, again, the scientific method comes of like approaching problems, being data driven, forming hypotheses, like actually challenging those hypotheses but the big picture has to be like so deep yeah and i think that was a great lesson i learned as an fde where and i think that is one of the most important things for me as a founder like you need to have um a balance of like confidence and conviction and humility and and i think that as again you learn that as an fde of like you are confident in

[25:17](https://realalpha.app/v/DwrYJCnoHZs/?t=1517) yourself and the things and you're helping this customer and you're solving problems together but you need to be you are talking they're the expert in whatever the thing is not you and so how do you work with them to solve problems together is um i think if you're a good fde that is the right the skill that you have yeah yeah um and i guess last question uh before we wrap um what have you found has been like maybe notoriously difficult to automate um in building like thread AI but even just like insights that you may have seen into industries yeah um I think they uh there are a lot of interesting ways um yeah I guess I think decision so decision making yeah yeah yeah um where um I'm not in the again and this is maybe also a hot take, I'm not in the AGI camp.

[26:11](https://realalpha.app/v/DwrYJCnoHZs/?t=1571) Oh, interesting. Yeah, yeah, yeah. You're on the AGI show. I believe in, like, intelligent systems and that, you know, we're building intelligent systems and using intelligent systems. You know, we can talk definitions all day, but I think a lot of the decision-making that I've seen attempted to be automated across the stack, across the process, be kind of unsuccessful except when the feature space is so constrained where you have a finite number of paths a finite number of opportunities whether it's like cool route to this person or hey do this thing or maybe we try this and be successful but when you like open up the aperture I've seen things struggle still but I think part of that is and I also used to say this also too past life of the death to innovation is also too this inability to get started

[27:17](https://realalpha.app/v/DwrYJCnoHZs/?t=1637) **Angela McNeal:** and AI workflows are always iterative so you're not going to have the perfect automated system

[27:22](https://realalpha.app/v/DwrYJCnoHZs/?t=1642) the perfect agent out of the get-go from the beginning you need to start and put in the building blocks and make sure that you're capturing the right information You're capturing that human AI exchange and then you're building this data asset that then your system can learn from and then get more intelligent and evolve over time. And so I think it's the expectation often that like, cool, I can automate every decision, everything from the beginning. Yeah. But in reality, you know, it's going to take a little bit. Yeah. Yeah. Well, I think that's a great place to wrap. Thanks so much, Angela, for talking to us about Thread AI.

[27:52](https://realalpha.app/v/DwrYJCnoHZs/?t=1672) Thanks for having me. Great to have you. Pleasure. A huge thanks to MTS sponsors. Autonomous Software Development for Enterprise Codebases Ship 5X faster. Blitzy.com. AdQuick. Make your brand a billboard. Out-of-home advertising as easy to scale as digital. AdQuick.com. Arena. Measuring AI performance in the real world. Arena.ai. Support for the show comes from VCX, the public ticker for private tech. The U.S. stock market started history's greatest wave of wealth creation. From factory workers in Detroit to farmers in Omaha, anyone could own a piece of the great American companies. But today, our most innovative companies are staying private longer, which means everyday Americans are missing out.

[28:35](https://realalpha.app/v/DwrYJCnoHZs/?t=1715) Until now. Introducing VCX, a public ticker for private tech. Visit getvcx.com for more info. That's getvcx.com. Carefully consider the investment material before investing, including objectives, risk, charges, and expenses. This and other information can be found in the Innovations Fund prospectus at getvcx.com. This is a paid sponsorship. Thank you.