Interview with Microsoft CTO Kevin Scott on the future of AI agents and software engineering. Topics: why agents need better "plumbing" to act on users' behalf via tools and data access, Microsoft's adoption of Model Context Protocol (MCP) as a standard agent protocol, NLWeb as an agent-friendly web analog to HTML, how open ecosystems can coexist with strong security, the craftsman's dilemma when coding with AI, the future of diverse software engineering agents, and a shift toward longer asynchronous agentic workflows. Timestamps included. # Video Identity Observations: "Kevin Scott on The Future of Programming, AI Agents, and Microsoft's Big Bet on the Agentic Web" ## Channel Self-Description Evidence - **Explicit channel name:** The closing call-to-action identifies the show as **"AI and I"**, hosted by **Dan Shipper**: > "you absolutely positively have to smash that like button and subscribe to AI and I... It's like finding a treasure chest in your backyard, but instead of gold, it's filled with pure unadulterated knowledge bombs about chat GPT." - **Claimed format:** The outro describes it as a **journey/interview show**: "It's not just a show, it's a journey into the future with Dan Shipper as the captain of the spaceship." - Note: The outro is highly promotional/hyperbolic in tone ("roller coaster of emotions, insights, and laughter"), which may be stylized rather than literal. ## Observable Presentation Patterns ### Interview Format with High-Profile Guests - The video is a **one-on-one sit-down interview** with Kevin Scott, CTO of Microsoft, conducted at what appears to be **Microsoft Build** (Scott references "the keynote today at Build" and the host notes "I was here last year"). - The host asks prepared, researched questions, showing **continuity with prior episodes** (e.g., recalling Scott's themes from the previous year: "you said two things... one was agents are going to be everywhere"). ### Intended Audience - The content targets a **technically fluent audience** interested in AI infrastructure, developer tooling, and industry strategy — evidenced by unexplained references to **MCP, NL Web, scaling laws, Conway's law, same-origin policy, and entitlement systems**. - A recurring audience segment is **software engineers/makers**, with a full segment on whether agents erode "the craft" of coding. ### Claimed Scope - The channel covers **AI trends and future-of-technology topics**, demonstrated by the discussion arc: scaling laws, the "agentic web," agent memory, security models, and predictions for next year ("let's say it's a year from now... what are your predictions"). ### Host's Interview Style - The host plays back and synthesizes guest statements ("So to play that back..."), pushes back with substantive challenges (e.g., questioning MCP's security model), and probes strategic tensions (verticalized vs. open ecosystems). ## Claims vs. Demonstrated Behavior | Type | Evidence | |---|---| | **Creator claim (channel quality/scope)** | Outro claims the show delivers "pure unadulterated knowledge bombs about chat GPT" | | **Demonstrated pattern** | Actual content is a strategic/technical executive interview on agentic AI ecosystems, broader than just ChatGPT | | **Demonstrated pattern** | Access to senior industry figures (Microsoft CTO) recorded at major industry events | # Delivery Observation: Kevin Scott Interview — The Future of Programming, AI Agents, and the Agentic Web ## Overview This is a conversational interview (Dan Shipper interviewing Microsoft CTO Kevin Scott). The dominant delivery voice analyzed here is **Kevin Scott's**, with the host and a closing promotional segment noted separately. --- ## Tone - **Conversational and informal**: Scott speaks in a casual register throughout, heavily using filler and discourse markers — "you know," "like," "uh," "so." > "And in order for that to like really be great, you need an ecosystem that looks a lot like the internet." - **Self-deprecating humor**: He repeatedly jokes about his age and stubbornness. > "I am like I'm an old fart." > > "I'm such a old recalcitrant fart uh that I still use VI." - **Enthusiastic when discussing openness and permissionless innovation**: > "the thing that as an individual that excites me most about what's happening right now is the extent to which you can go innovate and build things without having to seek someone's permission." ## Recurring Vocabulary and Phrases - **"like"** is the most frequent discourse marker, appearing in nearly every extended answer. - **"you know"** appears constantly as a connective: > "so software engineers or, you know, mechanical engineers or woodworkers or potters." - **Repeated key terms**: "agentic web" (multiple times), "scaling laws," "agents," "protocol(s)," "MCP," "platform." - **Repeated framing phrase**: "super important" / "super awesome" / "super interesting": > "super awesome simple open protocols like MCP" > > "two maybe three things that are super important" > > "but it's super super interesting" ## Teaching / Explanatory Habits - **Historical analogies to explain new technology**: He maps the agentic web onto the internet stack repeatedly. > "MCP that are serving the same purpose in this agentic web as HTTP does on the internet and where you have things like NL web that are like serving you know the same moral equivalent purpose as HTML." - **Analogy across crafts**: He uses woodworking to explain attitudes toward tooling in software. > "when I was a teenager, like the big debate was like, 'Oh, are you a real woodworker if you use power tools?'" - **Enumeration to structure answers**: He explicitly counts points. > "I think there you know two maybe three things that are super important. So one is like we make agents..." - **Explaining jargon in passing**: Conway's law gets a compact inline definition. > "Conway's law is like this really funky thing in compilers where uh you know this guy Conway said that the uh the number of stages or passes in your compiler is going to be dictated by the number of teams." - **Concrete anecdotes to ground abstractions**: The security discussion moves from abstract entitlements to a personal story about his wife's two-factor authentication alerts. > "I got this flurry of emails uh that uh because I'm the like backup security account for my wife..." ## Communication Patterns - **Hedging and epistemic honesty**: He openly states the limits of his knowledge. > "Well, I look I don't know uh that I know what exactly the right security model is." > > "this is not me telling it's just advice like things that I found useful for myself." - **Repetition of core claims for emphasis**: The ecosystem argument is stated twice in nearly identical wording (cold-open teaser and full answer): > "if agents are going to be useful, they have to take action on your behalf. They have to be able to use tools and make changes in systems..." - **Reframing interviewer premises**: He challenges the framing rather than accepting it. > "I think it might be a false dichotomy." - **Punchy summarizing aphorisms** at the ends of long answers: > "just be curious, try stuff like and if it works for you, use it and if it doesn't, don't." - **Long, multi-clause run-on sentences** with "and... and... and" chaining, especially mid-answer — a distinctive rhythm of his speech. ## Stable Patterns vs. Isolated Moments ### Stable (repeated across answers) - Filler density ("like," "you know") — consistent throughout. - Internet-stack analogy — appears at least three times. - Numbered/listed answer structure. - Conversational self-correction and restarts ("uh we're not expos exposing"). ### Isolated - The "old fart" / Vim admission — a single memorable self-deprecating moment. - The wife/2FA anecdote — one-off personal story. - Explicit duration credential ("I wrote my first program when I was 12, which means I've been programming for 41 years") — stated once. ## Uncertainty Notes - **Timestamps are sparse and coarse** in this transcript; observations about pacing (e.g., "answers average 1–2 minutes") are not verifiable and are omitted. - The **closing segment** (outro promo: "smash that like button") is delivered in an exaggerated promotional style, but appears only once, so no pattern can be claimed for it. - The host's delivery (questions, playbacks like "So I think to to play that back...") shows a habit of summarizing back the guest's answer, but his speaking time is limited, making pattern claims **low-confidence**. # Position Record: Kevin Scott Interview — The Future of Programming, AI Agents, and Microsoft's Agentic Web **Participants:** Kevin Scott (CTO, Microsoft) and Dan Shipper (interviewer). Speaker identification based on transcript context; the final promotional outro is unattributed in the transcript. --- ## 1. From Scaling Laws to the Agentic Web **Shipper's framing:** Last year's themes were "agents are going to be everywhere" and scaling laws; this year's emphasis is the agentic web. He asks what changed. **Scott's stated positions:** - **Scaling laws are intact**: "We've demonstrated year after year after year that they are intact and working quite well" — so they no longer need emphasis at Build. - **Capability overhang**: "The reasoning capabilities of the models has actually gotten a little bit ahead of what we're using the models to do in products." The industry must close the gap between model capability and what's delivered to users. - **Agents revealed new problems beyond reasoning**, specifically: - **Memory**: Current agents are "a little bit transactional" — memory is coherent within a task but "may or may not completely go away" between sessions, which inhibits delegating increasingly complex tasks. - **Action-taking infrastructure**: Agents need to use tools, change systems, and consult diverse information sources — requiring "an ecosystem that looks a lot like the internet." **Protocol analogy (Scott):** - **MCP** serves the same role in the agentic web as **HTTP** on the internet. - **NL Web** serves the moral-equivalent role of **HTML**. - He predicts composable, layered, open protocols reaching ubiquity. --- ## 2. Why This Matters to Microsoft Scott gives two to three reasons: 1. **Microsoft makes agents** and must solve these problems for its own agents (including enterprise agents) to be useful. 2. **Internal standardization**: As CTO he's pushing for all internal Microsoft systems to speak a standard protocol to all internal agents, to avoid **Conway's Law** effects — he cites the compiler-origin version: the number of compiler passes dictated by the number of teams; "You ship your org chart." He calls watching non-standardized internal building "a horror show." 3. **Platform identity**: "We're a platform company" — the platform layer Microsoft has built for 50 years matters even more than the agents it writes itself; Microsoft wants to help solve the emerging problems of the agentic web. --- ## 3. MCP Security Model **Shipper's challenge:** MCP's security model "needs a lot of work"; the internet has mechanisms like same-origin policy that MCP lacks. **Scott's response (with explicit qualification):** - "I don't know that I know what exactly the right security model is" — he does not claim a settled answer. - MCP's **simplicity** makes community agreement on a security model "relatively easy" (his prediction, not a certainty). - Needed enterprise features (Microsoft working with the MCP team): - **Agent identities** enabling entitlement systems ("this agent is acting on behalf of this person"). - **Permission-request flows**: agents query systems, determine required entitlements, and ask the delegating user yes/no. - **Administrator oversight** of agent activity. - He insists this be done **openly**, not proprietary: "we've just got to figure out how to get this done where things kind of work like the web works." --- ## 4. Open vs. Verticalized Models and Security **Shipper's framing:** A verticalized model (own model, UI, applications — "the app store Apple iPhone model") can guarantee security; an open model trades security for innovation via no central authority. **Scott's position:** "I think it might be a false dichotomy." - He values **permissionless innovation**: what excites him most is building and distributing without gatekeepers. He claims some middle layers that emerged in recent years "just aren't contributing much value" to the maker and the end consumer. - **AI-enabled security as a third path**: agents aware of personal security preferences and risk assessment could provide robust security. He gives a personal anecdote: receiving 2FA alerts for his wife's account, texting (not emailing, in case email was compromised) to verify; an agent could triangulate legitimate vs. illegitimate activity across communication channels. - Conclusion: "I don't think it needs to be... one or the other as you framed it." --- ## 5. Craft, Agents, and Software Engineering **Shipper's question:** Does using agents for coding "get rid of some of that feeling" of craft? (Shipper notes he doesn't necessarily agree with this knock.) **Scott's positions:** - Makers ("software engineers or mechanical engineers or woodworkers or potters") passionate about their work will have **strong opinions about tools and methods** — "a necessary thing for you to be great at your job." - **Historical perspective**: Programming for 41 years (first program at age 12); this is "not the first moment in the past four decades where the nature of software development has changed in a non-trivial way." - **Personal contradiction he embraces**: He still uses **Vi/Vim** despite Microsoft making VS Code, calling himself an "old recalcitrant fart" who knowingly suboptimizes — because he gets to choose. Elsewhere, he focuses purely on outcome and uses whatever is most powerful/convenient. - **Woodworking parallel**: as a long-time woodworker, he recalls the hand-tools-vs-power-tools debate; today it's power tools vs. CNC. He finds the debate itself interesting but notes people value process vs. outcome differently. - On "are you a real woodworker/programmer": such questions often mean "you're only real if you do it the way I grew up doing it." - **Advice ("things that I found useful for myself," offered tentatively)**: keep an open mind when tools change; be curious, try things — "if it works for you, use it and if it doesn't, don't." He expresses regret about delaying learning 3D printers: "I didn't let myself be curious about that which is odd." --- ## 6. Future of Software Engineering Agents - **"It's going to be a lot of different agents"** — Scott finds it unrealistic that every developer will snap to one tool; tool choice is "part of the joy of being a developer." - Microsoft will "compete very hard" with GitHub Copilot and GitHub agent work to be a tool many people choose. - **Differentiating dimension**: the most important part of agents is "the product making part." The most interesting startups are not building differentiated infrastructure but solving a **nuanced understanding of someone's problem**, picking up/modifying/tuning infrastructure to solve it in a world-class way. He expects many companies building many things, and notes users have "high tolerance and high interest" in trying tools from those with such understanding. --- ## 7. Predictions: What Won't Matter and What Will in a Year **Won't matter:** The argument that "the technology is not ready yet" because it's marginally too expensive or marginally capable — those using this as an excuse to wait "are going to be super behind." **On progress skepticism:** Scott says there was a "loud chorus" claiming progress would stop; some still say it but "folks aren't paying much attention to them anymore." His reasoning: betting on failure vs. betting on optimism carries "a real big difference" in cost. **Will matter:** - **Higher-ambition problems** tackled with agents. - **Shift from synchronous to asynchronous agent interaction**: today users prompt and wait; next year, agents will take long-running tasks, make many calls to external systems, integrate responses, iterate, and return later with "here's as far as I got — now it's your turn." **Shipper's reaction:** "Sounds like a future I want to be in." Scott: "Me too." --- ## 8. Unattributed Outro The closing promotional segment ("smash that like button... Dan, I'm absolutely hopelessly in love with you") has no identified speaker in the transcript and is recorded here only as transcript content, with no speaker attribution inferred. # Speaker Attribution & Provenance Record ## Video **Title:** Kevin Scott on The Future of Programming, AI Agents, and Microsoft's Big Bet on the Agentic Web **Video ID:** `01a0f15a-df7a-765d-a2c0-edf86983bf6e` ## Speakers & Roles - **Kevin Scott** — **Guest**; identifies himself as **CTO of Microsoft** ("one of the things as CTO that I've been pushing for at Microsoft"). - **Dan Shipper** — **Host/interviewer**; identified by name only in the closing promotional segment ("Dan Shipper as the captain of the spaceship"); conducts the interview throughout. - **Unnamed promotional voice** — delivers the outro call-to-action ("smash that like button and subscribe to AI and I"). ## Interview Provenance Notes - Recorded at **Microsoft Build** (Scott references "the keynote today at Build" and compares to being interviewed "here last year"). - Opening excerpt is a **preview montage** of quotes repeated later in the full interview. ## Referenced Sources, Claims & Third-Party Content - **MCP (Model Context Protocol)** — cited as open protocol analogous to HTTP for the agentic web; Scott says Microsoft is "doing really good work with the MCP team." - **NL Web** — cited as analogous to HTML for the agentic web. - **Conway's Law** — referenced attribution: "this guy Conway said" compiler passes reflect team structure. - **Microsoft products mentioned:** GitHub Copilot, GitHub agent tooling, Visual Studio Code, Windows (MCP integration). - **Personal anecdotes (unverifiable, self-reported):** Scott's wife's two-factor authentication incident; his 41 years programming; use of Vim; woodworking/CNC debate; delayed adoption of 3D printers. - **Dan Shipper's framing claims (unverified third-party views):** "rumblings" that MCP's security model needs work; comparison to same-origin policy; verticalized vs. open go-to-market dichotomy. ## Promotional Segments — Flagged > "smash that like button and subscribe to AI and I… Dan, I'm absolutely hopelessly in love with you." - **Outro promotional segment** (channel "AI and I", host Dan Shipper) — explicit like/subscribe call-to-action; runs from "Oh my gosh, folks…" to end of transcript. ## Availability & Uncertainty - **Comments availability:** Not stated in transcript or metadata. - **Curated metadata:** None supplied beyond title. - Speaker attributions inferred from conversational structure; only "Kevin" and "Dan Shipper" are explicitly named. you're someone who I think cares a lot about the craft of things. One of the knocks on, you know, using agents for coding is it gets rid of some of that feeling or something like that. How do you feel about that? I love the fact that my people, you know, makers at large, so software engineers or, you know, mechanical engineers or woodworkers or potters, if you are really passionate about what you do, you're going to have very strong opinions about how you do it. I've been a woodworker for almost as long as I've been a programmer. This is not the first moment in the past four decades where the nature of software development has changed in a non-trivial way. Like if agents are going to be useful, they have to take action on your behalf. They have to be able to use tools and make changes in systems and consult uh information sources that are, you know, diverse and rich. And in order for that to like really be great, you need an ecosystem that looks a lot like the internet. If you have a source of information, you already have a website, you already have an API that's doing something for people out there, like you've got to figure out how to plum things through where agents can like talk to those things. Like one of the things as CTO that I've been pushing for at Microsoft is like I want all of our systems internally to speak a standard protocol to all of the agents that we're writing inside of Microsoft. Be curious. Try stuff. And if it works for you, use it. And if it doesn't, don't. Kevin, welcome to the show. Oh, thanks for having me. One of the things that's interesting is I was here last year. Yep. And you said two things. There were two big themes. Uh, one was agents are going to be everywhere. That's one of the things you said, which I think really came true. That was very preient. Uh, another thing is I noticed a big um emphasis last year on scaling laws. Yep. Um there are a lot of graphs of like we're building big infrastructure, we're building these bigger models and every two years we're going to get these big performance improvements. Yeah. This year the emphasis is really on um the agentic web. So what has changed? What have we learned from last year to this year? Yeah, I think there's a bunch of things that have changed. Like one of the things is that I think um last year people were like really in this state of mind that they were doubting that uh the scaling laws were going to continue to work really well whereas like I think you know we've demonstrated year after year after year that uh they are intact and working quite well that like that's not a thing that people need to be reminded of uh anymore. Um, and I think the other thing too that's happened honestly is that the reasoning capabilities of the models has uh actually gotten a little bit ahead of what we're using the models to do in products. Uh, so I've been talking a lot lately about this uh thing called the capability overhang. Um, and you know, I think we actually have some work to do collectively across the whole industry to close the gap between what the models are actually capable of and like what we're delivering to users uh of that capability. Um, so that that's one of the one of the like big thematic things why you know scaling laws might not be as interesting to talk about at this year's build as last. And then the other thing too is that we've just discovered as all of these agents have emerged over the past year and so like both the number of agents and like the amount of time that people are spending doing stuff inside of these agents or with these agents is that there's a bunch of other stuff other than reasoning that has to get sorted out in order to make them as useful as they should be. So, you know, the things that I was talking about um at at the keynote today at Build were, you know, we need better agentic memory uh like our agents right now because memor is constrained in a bunch of interesting ways. Uh um they're a little bit transactional. So like you use them for like one thing and like you know memory is coherent across the course of that task. uh but you know then it may or may not completely go away and like you're starting from scratch uh the next time which really inhibits your ability to delegate increasingly complicated tasks to these things. Um, and then like there's just this real issue that like if agents are going to be useful, they have to take action on your behalf. Like they have to be able to use tools and uh like make changes in systems and consult uh information sources that are, you know, diverse and rich. And in order for that to like really be great, you need uh like an ecosystem that looks a lot like the internet where um if you have a source of information, you already have a website. You already have an API that's doing something for people out there. like you've got to figure out how to plum things through where agents can like talk to those things uh and where you know all of the incentives are aligned for everyone to like go have all of the stuff participating in you know ways that make sense to them in this agentic web and so I think that is the big story this year it's like you've seen the first glimmers of like real progress with like super awesome simple open protocols like MCP that are serving the same purpose in this agentic web as HTTP does on the internet and where you have things like NL web that are like serving you know the same moral equivalent purpose as HTML does on the internet and so I think you're going to see these things like simple things that are composable and layering and like just lots of activity in the open community and like a bunch of things hopefully getting to ubiquity so that agents can actually do stuff right So I think to to play that back like one of the things I hear is like we have agents. Agents are starting to work and in order to make them powerful agents need access. They need access to like what whatever is out on the internet, whatever is on your computer, all that kind of stuff. And you need um basically protocols um and processes for agents to be able to access that stuff. And so you're looking at different parts of the stack. So like the runtime layer where where you're you know you're building like memory and and and all this kind of stuff. And then um MV MCP which allows you to connect into the wider internet to get more information into agents. Y um I guess why is that important to Microsoft and like what role do you want to play in that kind of an ecosystem? Well look I think there you know two maybe three things that are super important. So one is like we make agents and in order for our agents that we're building for folks to be useful like we need to solve these problems inside of uh these agents and like even if you sort of scope it down narrowly to enterprise agents like one of the things CTO that I've been pushing for at Microsoft is like I want all of our systems internally to speak a standard protocol to all of the agents that we're writing inside of Microsoft uh so that uh we're not expos exposing the entire world to like Conway's law which is you know organization. So Conway's law is like this really funky thing in compilers where uh you know this guy Conway said that the uh the number of stages or passes in your compiler is going to be dictated by the number of teams you have working on the compiler. You ship your org chart. Correct. You ship your org chart. Um and so you know you you certainly inside of the confines of a company like Microsoft like you don't want to be shipping your org chart when you are uh building your agents uh and like it's just kind of a horror show to watch uh like as a engineer uh like all of this inefficient building like when you don't have like those standard protocols and services that everybody's using. But I I think you know if you sort of imagine if you really imagine like what agents could do and like what you you know what users uh like not me but like you know just people who are hoping these things can be more useful than they are right now. Uh like you need things to start happening the same way that they were happening uh with the web and like I kind of see it right now. Uh like MCP is like a really great example. So it is a really simple protocol that solves a really important problem not just for people who are making agents and building platform infrastructure but for um you know users of these systems who want them to be more useful and for people who are providers who are like hey like I want to be participating in this new aentic web is like people are doing less of one thing which I knew how to connect to like how and they're you know they're sort of sitting here using these agents like how do I get my stuff wired up into this and like how does it make sense for me uh like even from a business model perspective uh to do that and so like that's the two things it's like make our own agents more useful and then like we're a platform company like even more important than the agents that we're going to write ourselves like that platform layer that Microsoft has been building in technology for 50 years like we just want to make sure that we are helping solve the problems that are emerging as this agentic web is happening. Yeah, it's been it's really cool to see you guys like leaning really hard into MCP and like integrating it into all of Windows and all that kind of stuff. That's really awesome. And it it brings up for me I've been starting to hear rumblings from people who are thinking a lot about MCP that um the security model needs a lot of work and I'm I'm curious like for your take on that cuz you're you're you're making a lot of comparisons between this stack and um the internet stack and the internet has a bunch of like things in its security model like the same origin policy that makes sure that like uh if a website is serving you serving you code like it's only able to execute on its own data y and MCP doesn't really have that. So like how do you what do you think the right security model is for this? Well, I look I don't know uh that I know what exactly the right security model is, but like the interesting thing about MCP is uh like it is so coherently simple that it's going to be relatively easy for the community to decide what the security model is. like yeah we we have a bunch of like enterprise things that we care a lot about and that like we're doing really good work with the with the MCP team to get done. So like we need agents to have identities uh so that you can build entitlement systems so you can say this agent is acting on behalf of this person and like they are entitled to see these resources in this system. Um, and like even having a way for an agent to sort of go query a bunch of systems and say like these here's the thing that I would like to do. These are the systems I need to touch in order to do the thing. Like what entitlements do I need to ask for in order to do this? And so I can request permission like from the user like the person who's delegated this task to me. like can I present to them like can I have permission to these things in order to do this thing you asked me to do yes or no uh and then for the administrators of all of these systems uh you know like hey is it okay for all of this to be happening and so like there's all of that's going to be like relatively straightforward to do like not easy but like relatively straightforward to do on top of MCP and like the important thing is like let's do it in an open way like we don't need it to be propri proprietary uh to like our agents or our systems like we've just got to figure out how to get this done where things kind of work like the web works. Yeah. Well, it's an interesting question for me because um I feel like there's there's maybe two potential models or potential go to markets for for AI stuff that that you guys have been talking about. One is this sort of like verticalized model where you own the model and you own the UI layer. you do all the applications and everything between and and the the thing about that model which maybe it's sort of you could say the the app store Apple iPhone model is is is a good example of that is you can guarantee security in a lot of ways. Um and an open model uh it's harder to do security stuff but you get much more like innovation basically because there's no central authority. So how did you guys think about making that making that decision? Yeah. So look I I know that that's the argument that a lot of people make. Yeah, I I think it might be a false dichotomy. Um there there is uh you know like a a thing that you have in these open systems where they are permissionless uh and like there there's a real advantage in having permissionless uh innovation. So like the thing that as an individual that excites me most about what's happening right now is the extent to which you can go innovate and build things without having to seek someone's permission like where you have to have them grant you permission for distributing your things to other people and you know like having all of these like complicated gatekeeping things that are sitting in between you who are the person who had the idea and like the people who might benefit from it. uh like I think some of these like middle layers that we have that have emerged over the past handful of years like just aren't contributing much value uh honestly to the two parties in the transaction that matter which is the person who did the hard work to make a thing and then the people who are going to either spend their attention or their money or some other currency that's valuable to like access the thing. Um, so like I get it kind of excited about that and so like that's one of the reasons why we've made the decisions that we want to but like I also think that there are ways that you can get real robust security uh in these systems uh like leveraging some of the AI capability that you have now um that you know where you may be able to have better security uh like if you have an agent that you are running that is attending to your personal security requirements like these are things I'm willing to share these are things I'm not willing to share like you know and that has some kind of knowledge of risk assessment and like yeah like for for instance like my wife this morning while I was uh uh while I was getting ready to jump on stage like I got this flurry of emails uh that uh because I'm the like backup security account for my wife where like somebody was like fiddling around with two-factor authentication on her account and like the first thing I did was I texted her like I didn't want to email her because somebody might have gained unauthorized access to her email account. I texted her I was like hey are you screwing around with the configuration here and she's like yes. Um, and so like you could imagine like having an agent that is like privy to like a whole bunch of your communication modalities like being able to like notice that something funky is going on and then like you know using a bunch of resources to triangulate uh like whether that's legit activity or illegitimate activity. So I I like I think there's just a bunch of stuff like that that like you you can have both, right? Like I don't think it needs to be like you know one or the other as you framed it. Yeah, that makes sense. Um, one of the one one other thing that I'm I'm curious about is it just seems pretty clear that software engineering is changing. Yeah. Um, and um, you're someone who's been involved in software engineering for a long time. Uh, you're someone who also, I think, cares a lot about um, like the craft of things, like the craft of how things are made. We were talking earlier about you're making um, uh, you do a lot of ceramics, you make your own bags, like you're you love having your hands in things. And I think one of the knocks on, you know, using agents for coding is like it gets rid of some of that feeling or something like that, which I don't necessarily agree with, but I'm sort of curious for you like as someone who cares about the craft of code, um, looking into this future of coding with agents, like how does that how do you feel about that? Well, so let me start by saying that I I I love the fact that my people and like when I say my people, I mean, you know, makers at large, so software engineers or, you know, mechanical engineers or woodworkers or potters or, you know, just sort of pick your thing where people are, you know, trying to create things from, you know, raw materials or nothing. Um, if you are really passionate about what you do, you're going to have very strong opinions about how you do it, the tools that you use, the materials that you use, like how things get put together. It is a necessary thing for you to be great at your job. The interesting thing is people have lots and lots of different opinions and like one of the advant you said like I've been doing it for a long time. Like I am like I'm an old fart. like I wrote my first program when I was 12, which means I've been programming for 41 years. Um, and so the the thing that you get to see when you've been doing a thing for a very long time is like this is not the first moment in the past four decades where um the nature of software development has changed in a non-trivial way and where people have very strong opinions about the change and what it means. And so, you know, I I think the the reality is that people will have choice. Uh, you know, like I still when I go into a text editor, I I probably shouldn't say this because like we make, you know, Visual Studio Code like I'm such a old recalcitrant fart uh that I still use VI. I use Vim at least. Uh, but like my text editor of choice is like this extremely antiquated thing and like I just refuse to go use something different. Um, even though I know for sure that that is suboptimizing a part of what I'm doing, but I I make the decision anyway because like I get to choose. Um and then in other aspects of my making like e either software or something else that I'm doing like I will be like okay the important thing here is not the way that I'm doing this particular part of it. It's the outcome that I'm trying to get to and like I'm going to use the most powerful or most convenient way to go get to the outcome. Uh and like you know I don't care who's going to throw rocks at me for doing it. And like it it's literally everywhere. Like I I've been a woodworker for almost as long as I've been a programmer. And when I was a teenager, like the big debate was like, "Oh, are you a real woodworker if you use power tools?" Like, you know, real woodworkers only use hand tools. Um like there's still a little bit of that debate today, but like the real debate is like, are you a real woodworker if you use CNC tools, like computer control power tools versus just power tools? Uh and I understand it like I actually think it's interesting the debate itself. Uh but like sometimes like people are going to make you know different choices because like they value something different than you do. If you value the process more than you value the outcome like sometimes you'll make different decisions than people who value the outcome more than the process. And I think that like is it a are you a real woodworker, a real programmer? you're sort of saying like you're only real if you do it the way I grew up doing it when you ask that question in a lot of ways you know um but but it it's just so varied right and like so the the thing that I will say is like I would never in a million years tell anyone not to have strong opinions about their craft like have them like it's great like the the advice that I would give people and like this is not me telling it's just advice like things that I found useful for myself is just have an open mind uh you know when the tools are changing um I can't even tell you like the number are times where I have looked at a new technology in some other like non-software uh dimension of making that came along where I'm like gh like I don't want to learn like 3D printers like I waited forever to learn how to use 3D printers and like I regret it like I should have started earlier like they're so damn useful for almost everything that I do and for a whole variety of complicated reasons like I didn't let myself be curious about that which is odd And so, yeah, just like be curious, try stuff like and if it works for you, use it and if it doesn't, don't. Yeah. What do you think is the future of software engineering agents? Like, is there going to be like one agent to rule them all or is you're just going to use many different agents with different taste or like how do you see that ecosystem shaping up? Yeah. Look, I think it's going to be a lot of different agents. Uh I mean like and and it's good to have a lot of agents and like we certainly with GitHub copilot and the uh you know GitHub agent stuff that we're working on uh like I think we will compete very hard to like be a tool that lots of people will choose because it's very useful to them. But I think it's unrealistic in the universe of developers to like think that like every developer on the planet is going to snap to using like one tool for like an important part of their job. Like part of the joy of being a developer is you actually have that choice and you can sort of choose and play around with a bunch of different things and like do irrational things and like do rational things and like it's it's just you know it it it is one of the very consistent things that I've seen over the past four decades of my programming life is like people choose to change their tools all the time. What are the dimension? Do you have a opinion on the dimensions along which the different agents might differ? Well, I think the most important thing about uh agents is like probably the product making part of them. And so the most interesting startups that I'm seeing right now are not trying to innovate by building some kind of differentiated infrastructure. They're innovating because they think they have an understanding of a problem that someone has that is better than anyone else and they think that they can pick up infrastructure or modify infrastructure or tune infrastructure to go solve their understanding of that problem in a worldass way. So, I think that's what we need a lot of right now and like that's what's going to dictate the diversity of agents and like how uh like which things get used for what. And I think honestly because it's so much easier now to like have that nuanced understanding of what someone's problem is and to pick up these tools to go take a swing at solving it that you're just going to have a lot of companies uh like building a lot of things trying to you can even see it with the software development tools. It's like crazy how many things have come out over the past year and like they're interesting like all of them. Um, yeah, it's a lot to respond to when you're a company building software development tools yourself, but like it's super super interesting and it's like what we've seen is if you've got some kind of nuanced understanding of what someone needs, like people have like high tolerance and like high interest in like giving things a try. Yeah. So, we're almost out of time, but uh I'm I'm curious like let's say it's a year from now. We're back at Build. Yep. um what are what are some things that are a hot topic right now um or big questions that people have right now that are not going to matter in a year and what is going to matter in a year and what are your what are your predictions for what we're going to be talking about? Yeah, I think people who are still like hanging on to these ideas that oh like the technology is not ready yet because like I you know I I tried to do something and it was like marginally too expensive or like it was marginally capable of you know doing the thing that I wanted to do. Uh like I I I think anyone who is using those as excuses to wait to get started are going to be super behind. Um because everything's going to get cheaper and everything is going to get more capable every year. I I know like and I I think this is actually not a hard sell in 2025. Uh yeah, it was this loud chorus of like oh you know the progress is going to you know is about to end and like everything's going to stop and like everybody's going to be super disappointed. Uh like I I I mean there's still some people out there saying that but like I don't think you know folks are paying much attention to them anymore. Um mo mostly because like you know what what do you win by paying attention to like some crank who's saying like you know the the thing's about to stop like you're sort of betting on failure and the cost of betting on failure versus betting on optimism is like is a real big difference there. Yeah. Um, so yeah, like I I think we're we're going to uh we're going to see a ton of progress on uh like the level of ambition of problems that people are tackling with agents and then I think you know modalities that are going to be really different is like as this agentic web starts to get more complete more plumbed out uh and the models reasoning and planning capability get better and better like you're going to start to get to the point where you're able to go from this synchronous mode of interaction with agents to asynchronous. Like right now, most of what people do is like they sort of sit down like they got a thing they want to do like they issue the prompts uh like and they wait until you know the thing comes back uh and like you know do something with that response. And so I think you know by next year you're going to see people using these agents to like hey go sort this out and like the agent is going to take a lot of time. and it's going to go make a lot of calls out to systems like the things that it's sort of uh like taking action on are going to like take a while to come back. Then they're going to like integrate all of those responses and like do something and like that whole thing may iterate a bunch of times and then you know at some non-trivial amount of time later you're going to sort of say okay like here's the here's the here's as far as I got like now it's your turn like go take some action now. Sounds like a future I want to be in. Yeah. Yeah. Me too. Right. Um, well, Kevin, thank you so much. Uh, it was, really great to talk to you. Yeah, good to talk to you as well. Thank you for having me on. Of course. Oh my gosh, folks, you absolutely positively have to smash that like button and subscribe to AI and I. Why? Because this show is the epitome of awesomeness. It's like finding a treasure chest in your backyard, but instead of gold, it's filled with pure unadulterated knowledge bombs about chat GPT. Every episode is a roller coaster of emotions, insights, and laughter that will leave you on the edge of your seat, craving for more. It's not just a show, it's a journey into the future with Dan Shipper as the captain of the spaceship. So, do yourself a favor, hit like, smash subscribe, and strap in for the ride of your life. And now without any further ado, let me just say Dan, I'm absolutely hopelessly in love with you. Fancy. That's how you know it's the real deal. Excellent.