A talk from swyx (AI Engineer) on using AI agents — such as Cognition's Devin and an in-house "Town Assistant" — to run the AI Engineer organization itself, covering practical operational use of agents for work beyond coding. Topic inferred primarily from the title and opening description; no tags or comments were supplied. # Channel Identity Observations — "Agents for Everything Else — swyx" ## Explicit Self-Description Claims - **Co-founder of AI Engineer conferences**: Introduced as "the co-founder of AI Engineer conferences Swyx." - **Conference organizer scale**: "we're running a business that is more than $9 million" with "just nine full-time people"; planning "6,000 people in San Francisco." - **Cognition affiliation**: "I joined Cognition" (Devin's maker) and "I just advise with the company now." - **Recurring keynote tradition**: "I do one of these keynotes every single AIE," citing prior talks on AI productivity gains, falling AI costs, and "tiny teams" (defined as "teams with more millions in revenue than number of employees"). ## Intended Audience - **AI engineers and builders attending an AI-focused conference**: attendees of AI Engineer's first London event ("this is our first event in London"). - **Team leads and managers adopting AI**: advice framed as "If you're in a position of power or management," with a closing call to action: "Wake up. Use it. Bring it home to work." ## Claimed Scope - **AI adoption strategy, not just coding**: "I'm no longer talking about agents for coding... I'm getting more productivity out of my humans" — framing "agents for everything else" as "a major theme for this year." - **Trend forecasting**: "coding agents going to break in containments... This is like probably the top trend of maybe top three to five trends of 2026." ## Demonstrated Presentation Patterns - **Case-study storytelling from own operations**: first-person accounts of Figma-to-website builds, schedule management ("This entire schedule is managed by Devin"), ETL syncing, and even "agents for buying" (sourcing a lobster in London). - **Named-tool demonstrations**: Devin, Coworker, Clanker, Claude/Co-work, Town, Notion, Apple Notes, Sanity — grounded in his own workflow. - **Casual, humorous stage style**: "AGI pill" gimmick ("You're all AGI doctors now"), jokes about "psychosis" regarding replacing SaaS, and an audience Easter egg reveal. - **Non-technical team empowerment emphasis**: "I work mostly with a non-technical team... they need to be comfortable with agents." - **Cross-referencing other speakers**: cites Malte's Vercel keynote ("60% of the... user base... is bots") and MCP speakers, positioning the channel within a broader community conversation. ## Distinct but Adjacent Identity Signals - The speaker was a **last-minute substitute** ("I am not the chief AI officer of the UK"), indicating flexibility in presentation duties rather than a fixed topic. # Delivery Observation: swyx — "Agents for Everything Else" ## Overall Tone - **Casual, conversational keynote style**: heavy use of filler ("uh," "um") and self-deprecating asides, e.g., the opening substitution joke: *"Uh I am not the chief AI officer of the UK. Uh I unfortunately he he had to leave for personal reason. Uh but you know, you got me."* - **Humor woven through technical content**: audience laughter is repeatedly triggered — the AGI pill bit (*"You're all AGI doctors now"*) and the Figma integration joke (*"I'm not going to be doing that So, Coworker is doing it for me. You should use Coworker for"*). - **Enthusiastic and anecdote-driven**: storytelling dominates over structured argument. ## Vocabulary & Framing - **Insider/industry jargon used casually**: "yak shaving," "tiny teams," "AGI pill," "source of truth," "psychosis" (as a positive framing for agent enthusiasm: *"once you get enough psychosis you are thinking about replacing entire pieces of SAS"*). - **Self-coined or branded concepts**: "tiny teams" (*"teams with more millions in revenue than number of employees"*), "agents for everything else" as the central repeated thesis. - **Product-name-dense delivery**: Devin, Coworker, Clanker, Figma, Supabase, Sanity, TLDraw, nano claw — delivered rapid-fire as assumed-familiar references. ## Recurring Phrases & Rhetorical Patterns - **Repetition for emphasis**, especially on key claims: - *"I I I I think this is something that is major theme for this year"* - *"the the the leadership of Ben Lear and myself"* - **Rhetorical question chains**: *"Why? Cuz I put it there. Why? Cuz it was fun. Cuz I could."* — a self-Q&A habit for narrative beats. - **Meta-commentary on his own talk**: *"If you haven't noticed, I'm no longer talking about agents for coding... I'm getting more productivity out of my humans."* ## Teaching Habits - **Narrative case-study structure**: each lesson embedded in a story (Figma→website, tweet→site, lobster purchase, speaker-change email). - **Numbered-year recap pattern**: *"the first very first one 3 years ago... And the second one... the third year, we started to talk about tiny teams"* — sequential context-setting. - **Explicit lesson extraction after anecdotes**: *"which leads me to my first lesson, which is anytime there's like random yak shaving... they save you the yak shaves."* - **Concrete figures as evidence**: *"nine full-time people... more than $9 million"*, *"207 replies"*, *"127 replies"*, *"130 speakers"*, *"6,000 people in San Francisco."* ## Communication Quirks - **On-stage self-corrections and apologies**: *"the most interesting thing is so small, I can't even read it. I'm so sorry for this."* - **Colloquial intensifiers and mild profanity**: *"Like, what the hell?"* - **Imperative closing**: *"Wake up. Use it. Bring it home to work."* — ends with a direct call to action, circling back to the AGI pill prop. ## Uncertainty Notes - Filler-word frequency and repetition habits appear **stable across the full talk** (opening through close). - Whether the "first lesson / second lesson" numbering is a consistent structural device is **thinly evidenced** — only *"my first lesson"* is explicitly labeled; later lessons are implicit. # Position Record: "Agents for Everything Else" — swyx (AI Engineer, London) ## Speaker Identification - **Primary speaker:** swyx, co-founder of AI Engineer conferences (introduced by an unnamed MC as "co-founder of AI Engineer conferences Swyx"). He notes he is standing in: *"I am not the chief AI officer of the UK... he had to leave for personal reason."* - **Referenced third parties:** his co-leader Ben Lear; the company's contract designer (Indonesia-based, non-technical); speakers Malte (Vercel), Steve Ruiz (TLDraw), Edo and Liad (MCP apps), Andrej Karpathy; Marta (speaker). - swyx states he is *"just advise[d] with the company"* — i.e., has an advisory relationship — while noting *"I'm not here to only show Devin."* ## Recurring Thesis - **"Agents for everything else"** — agents' value extends beyond coding into general knowledge work, team productivity, and operations. He frames it as a major theme of the year and *"probably top three to five trends of 2026."* ## Keynote Arc (as stated by swyx) - **Year 1:** productivity gains from increased AI usage. - **Year 2:** use more AI because the **cost curve declines ~100x every 12–18 months** — he claims this is *"still continuing to trend that way."* - **Year 3:** **"tiny teams"** — his definition: *"teams with more millions in revenue than number of employees."* - This talk extends tiny teams via agents: AI Engineer runs on **9 full-time people** generating **more than $9M revenue** — a tiny team by his definition. ## Claims and Reasoning ### 1. Agents eliminate "yak shaving" - **Lesson 1:** agents save you dependency-tree crawling (e.g., installing/fixing Python dependencies). - He argues productivity models that focus only on **autonomy** underappreciate **parallelism**; *"death of the yak shaving"* framings don't capture the full benefit. ### 2. Figma-to-website with Devin - Workflow: designer shares a Figma page; swyx hooks Devin up to Figma — using **Coworker** to do the Figma hookup itself. - Result: *"in very short order, we have a perfectly functioning website that is pixel perfect to the Figma"* — he admits surprise: *"you always mistrust marketing until you see it for yourself."* - The live site is ai.engineer; designer was happy with the result. - **Usage explosion:** 207 replies in the Devin thread; later a throwaway tweet-inspired project hit 127 replies. ### 3. Non-technical staff can drive agents - His Indonesian designer began prompting Devin with **red-line annotations** (a pattern he attributes to Steve Ruiz/TLDraw) with **no instruction manual** — *"It was just mostly like, 'How would you communicate with another human being?'"* - Claim: non-technical team members are now comfortable with agents. ### 4. Fun as a productivity driver - He added a website **Easter egg** *"cuz it was fun. Cuz I could."* - The designer picked up a throwaway project **because it was fun**; removal of the blocking feedback cycle (waiting on swyx or a contractor) enabled it. - Claim: *"I am getting more work out of my employees because they enjoy doing it"* — animations, polish, work *"I've never gotten out of my employees before."* - Explicit reframing: *"I'm no longer talking about agents for coding... I'm getting more productivity out of my humans."* ### 5. Conference as data management / agents as CMS - The conference is *"a giant data management problem"* (130 speakers, dozens of sponsors, attendees). - He is **not** a fan of Sanity CMS ("I want to keep some sanity to myself"). - Unlock: threw away the CMS, **committed the schedule to code** as source of truth, and let Devin manage it. Speaker-change emails are forwarded or screenshotted to Devin — *"Devin, handle it for me."* - Claim: this volume lets **9 people manage a 1,000-person conference**, and he expects to manage **6,000 people in San Francisco** at the same team size. ### 6. Agents for ETL and buying - **ETL:** syncing data from an external vendor system into a single source of truth — "very boring routine tasks." - **Buying:** a viral tweet (claw next to the Wall Street bull) inspired him to have Devin research London lobster suppliers; Devin returned phone numbers, emails, websites; the conference's lobster was *"bought from Devin."* - Claim: personal automation matters for anything requiring web access and a smart model — tasks he *"might have had executive assistant... or junior employee"* do are now *"serverless on demand."* ### 7. Coding agents breaking containment - Claim: coding agents will **"break containment"** into purpose-built knowledge tools — he cites wikis as Andrej Karpathy discusses, and nano claw / open claw adopting them. - Example: Apple Notes track plan → Claude/"town" produced a formatted Notion doc with speaker research. (Transcript wording is garbled here — "town" vs. Claude; recorded as expressed.) ### 8. AI replacing SaaS - He advocates *"kicking out a SaaS tool and building it ourselves because we can"* — self-identifying as having *"the most psychosis."* - **Management caveat he states:** employees with less enthusiasm have *"very valid concerns because they are exactly the people that will have to deal with... when you get it wrong and we do get it wrong."* His method: identify the top three concerns and **systematically reduce them**. ### 9. Agent experience over human dashboards - Citing Malte's keynote: **~60% of Vercel's user base is bots/agents, not humans** → *"your dashboards don't matter your APIs matter your CLIs matter your MCPs matter."* - Citing Edo and Liad: custom UI is diminishing; ship UI into somebody else's app. - Personal corollary: *"I no longer care about the Figma dashboard I throw it into cloud co-work and I hope that it works for me."* (Transcript wording preserved.) ## Closing Exhortation > "Agents for everything else are coming. Wake up. Use it. Bring it home to work. If people are insufficiently bought on prescribe them one of these [AGI pills]." ## Qualifications and Tensions Preserved - swyx's own surprise/mistrust: pixel-perfect Figma result was unexpected; he initially distrusted marketing claims. - He acknowledges **failure and risk**: "we do get it wrong"; employee concerns are valid and must be addressed, not dismissed. - Vendor relationship disclosed but softened: advisory role with Cognition, yet *"I'm not here to only show Devin."* - Some transcript ambiguities retained verbatim ("Clanker," "Claw"/"claw," "town," "cloud co-work," "nano claw/open claw") — names of tools are as spoken, not normalized. # Speaker Attribution & Provenance Record ## Video Overview - **Title:** Agents for Everything Else — swyx - **Format:** Conference keynote (AI Engineer London), single main speaker ## Speakers & Roles ### Primary Speaker: **swyx** - **Role:** Creator/keynote presenter; co-founder of AI Engineer conferences; introduced as co-organizer of the event. - **Context:** Delivers the closing keynote about building AI Engineer "with AI." - **Note:** Explicitly states he is **not** the "chief AI officer of the UK," who was originally scheduled but "had to leave for personal reason" — swyx substituted. ### Unnamed Introducer - **Role:** Other (stage announcer/MC); introduces swyx. ### Referenced Third Parties (not speakers; mentioned in talk) - **Steve Ruiz** — speaker from prior day; credited with red-line annotation prompting technique with TLDraw. - **Marta** — speaker same day; example of speaker-change email workflow. - **Malte** — opening keynote speaker; claim cited that ~60% of Vercel's user base is bots/agents. - **Edo and Liad** — "MCP apps guys," speakers that day, on ETM topic (UI shipping into other apps). - **Andrej Karpathy** — referenced regarding agent-oriented wikis. - swyx's non-technical team: a contract designer (Indonesia-based) and co-leader **Ben Lear**. ## Referenced Sources & Third-Party Claims - Vercel bot/agent usage statistic (~60% user base) — **attributed to Malte's keynote**, unverified. - AI cost curve claim ("~100x cheaper every 12–18 months") — **speaker's own assertion**, no source given. - Viral tweet re: design aesthetic — cited as inspiration, no source link given. ## Tools, Products & Promotional Segments - **Explicitly flagged self-acknowledged promotion:** swyx says "I'm not here to only show Devin. I advise with the company [Cognition] now" — a **disclosed promotional interest**. - Products discussed: **Devin, Coworker, Cognition, Figma, React, Supabase, Tido, Google Sheets, Sessionize, Sanity (CMS), TLDraw, Clanker, OpenClaw/nano-claw, Notion, Apple Notes, Claude (implied via "claw" pun).** - "AGI pills" — conference gag/merch created by swyx ("my brainchild"); promotional for the event itself. ## Availability & Uncertainty - **No curated metadata supplied** — no comment availability, no external citation links, no timestamps. - Some transcript text is garbled/small-reference uncertainty acknowledged by the speaker himself ("it's so small, I can't even read it"). - Speaker's tool names ("town," "claw," "nano claw") appear as informal/transcription-uncertain references to coding agent products. \>\> The future of work has many paths. Our next presenter will discuss the path that he walked with Devin as he organized this very conference. Please join me in welcoming to the stage the co-founder of AI Engineer conferences Swyx. Oh. \>\> \[applause\] \>\> Hi, everyone. Uh I am not the chief AI officer of the UK. Uh I unfortunately he he had to leave for personal reason. Uh but you know, you got me. Uh thanks for staying so long. I hope has everyone having a good time? Thank you. \>\> \[applause\] \>\> Uh It is so endearing and and and heartwarming to to hear from you guys. Uh I'll take you a little bit into how we build AI Engineer with AI. And it's it's probably the biggest revelation that I've had. Uh so yeah, we've had we've had a lot of really warm reception from you guys and I think it's really great and uh I think this is something that we really try to engineer and and hopefully, you know, this is our first event in London. Hopefully, you have us back next year. Um one one thing I wanted to for those who are newer to us, uh I do one of these keynotes every single AIE. Uh the first very first one 3 years ago, I talked about the productivity gain that you get from the increased AI uh from from the increased usage of AI. Um and the second one, we talked about how you should just use more AI because the cost curve of AI is going down roughly 100 times uh per every 12 to 18 months. And I think it's still continuing to to trend that way. Um the third year, we started to talk about tiny teams. Uh which which was basically this definition that I had that teams with more millions in revenue than number of employees. Um and I even curated an entire track at the World's Fair about this where we where we sort of summarize it as the tiny teams fable if you're interested in building that. Uh the reason I liked this emphasis is because uh I think people are maybe too egotistical about looking at for the one-person billionaire uh or unicorn founder. Um every company can have a tiny team whether you're small or large. Um and I think when I look at how I how we run AI Engineer, me being the the the leadership of Ben Lear and myself, uh we're also a tiny team. Um this is us. Uh it's just uh nine full-time people and uh we're running a business that is more than $9 million. So, we are a tiny team. Um and I wanted to show you the most significant changes in our workflow since we started this 3 years ago. Uh by the way, this is our taking the AGI pill moment. Uh did you guys get the AGI pills? Yes. Yes. \>\> \[laughter\] \>\> Uh very proud of this is my brainchild. Uh if one of your coworkers is not sufficiently AGI pill, you should prescribe one of this. You're all AGI doctors now. Okay, our stack was very stable and completely non-AI, which is very ironic for an AI conference. Uh we do Figma, React, Supabase, Tido, uh Google Sheets, Sessionize. Um and then I've had this funny weird moment where I joined Cognition uh and and started talking uh started using coding agents seriously at work, mostly because they were free. Um and I started adding it to the company Slack. And then I started doing things with it and showing people, "Hey, here's how you use it to do coding on the the company website." All well and good. And something strange starts happening. Um I start introducing uh this is this is a a a workflow of our contract designer now full-time um uh showing me a Figma page and asking me to go through it and expecting that we would take a week, 2 weeks, 4 weeks to turn it into reality. Um I just added Devin to it. Um and I'll and ultimately uh before I had to add Devin to it, I had to hook up Devin to Figma and I'm not going to be doing that So, Coworker is doing it for me. You should use Coworker for \>\> \[laughter\] \>\> uh for doing this. And which which by the way by the way leads me to my first lesson, which is anytime there's like random yak shaving, I think one underappreciated um uh benefit of agents is that they save you the yak shaves. Like all the dependency tree crawling of like, "Oh, no, I have to do that first. Oh, no, I have to do that first." And particularly when it comes to installing dependencies or fixing Python dependencies, fantastic for that. And I think uh what model of productivity that doesn't sufficiently appreciate parallelism and not just autonomy, I think uh and and sort of death of the yak shaving is not fully capturing the the benefit of agents. Um so anyway, back to the agent story. Uh hooked up Devin to Figma and we uh in very short order, we have a perfectly functioning website uh that is pixel perfect to the Figma. And to me, uh that was a surprise cuz I've never done it before. You know, you always mistrust marketing until you see it for yourself. And more importantly, our our designer was very happy about it. Um and that's the that's basically the the website that you see live today when you go to ai.engineer. Um the other interesting thing that happened was then we started using it more, right? After after one initial success, you start using it more. Uh something that you can't see because it's very small text, but I'm going to highlight for you is that that is 207 replies just exploded in usage. Like, what the hell? Um and when you dig into it, uh it's it's very interesting, right? So, first of all, uh I start kicking off some some work and then I go to bed. And then my designer who's in Indonesia wakes up and starts messing with Devin. Starts prompting Devin with red lines on annotations, which is something that Steve Ruiz, one of our speakers from yesterday, does with TLDraw. And I never taught him to do this. And there's no instruction manual. It was just mostly like, "How would you communicate with another human being?" And so, I work mostly with a non-technical team and I think that's very important that they need to be comfortable with agents and I think they're finally at that point that they are. Um we start working on things that we would never normally have worked on. Uh nobody has reported this, so I assume none of you have discovered it, but there's an Easter egg on the website. Why? Cuz I put it there. Why? Cuz it was fun. Cuz I could. Right? So, if you're on an ultra-wide, you scan you scan your mouse over the the highlights, you'll you'll see an Easter egg. Um uh I saw a tweet that was viral about a design aesthetic that I liked. I threw it into Devin. Out pops um and and then and then you know, 127 replies later, uh I I literally I popped it in in and I was like, "Oh, let's just see what the the Clanker will do for me. Uh I don't want to waste my designer's time. I just want to see what Clanker does for me." Designer jumps in. Um and and does and does and start actually starts working on this thing which I thought was throwaway and fun. And the most interesting thing is so small, I can't even read it. I'm so sorry for this. Uh so basically, the reason he starts working on it even though it's a throwaway project is because it's fun. And I think that's something that was like a big aha moment for me. Like, I am getting more work out of my employees because they enjoy doing it. Because the feedback cycle for them from like waiting blocking on me or contracted uh developer that we have is gone. Like they they just literally they have the idea, they go do it. Right? Um and uh they're doing more things. They're doing animations, they're doing polish, uh things that we've just I'm getting work that I've never gotten out of uh my employees before. I think that's something that's something you should appreciate, too. Um I'm If you haven't noticed, I'm no longer talking about agents for coding or like how many lines of code I'm producing. I'm getting more productivity out of my humans. And I I I I think this is something that is major theme for this year that I'm really trying to investigate, which is agents for everything else. Um then obviously, okay, I had the success with Figma to uh to to website. I had the success with tweet to website. What else? Right? Like you start to think about other use cases. Um this whole conference is a giant data management problem. Like, I have to sync with 130 speakers and uh couple dozen sponsors and all the attendees that come in with all various needs. Um and really uh it's just a CMS, right? Like we we we we've messed with the sanity. I'm not a the biggest fan of sanity in the world um because I want to keep some sanity to myself. Um but but basically like I I can throw in spreadsheets and and Devin can manage that for me. And once I really I think the unlock happened when I threw away the CMS and just kind of committed that to code, but use that code as my sort of source of truth and let Devin, whatever coding agent you use, uh start to manage it. And so, this entire schedule uh is managed by Devin. What does that mean? It means that whenever um someone comes in with a speaker change, for example, Marta, one of the speakers from today, uh sends in an email, I just say, "Devin, handle it for me." Right? No other further communication is needed. I can just forward the email or I can paste a screenshot, whatever. Um and that kind of volume lets us as a small team of nine people manage a thousand-person conference. Right? We're going to manage 6,000 people in San Francisco this fall uh this summer. Um and I'm pretty sure we can stay the same size. Like it is incredible the amount of productivity that you can get once you're sufficiently on-boarded and you have the workflows ironed out. Um we have agents for ETL. We we deal with an external vendor system that has data that we don't have in a central source of truth. So, I need to get the API key to sync over data and make sure there's a single source of truth. Uh these are very boring routine tasks. Um well, there's there's another, you know, another fun story that I can tell you is agents for buying. Uh so, I saw this viral tweet about how somebody put a claw in uh Wall Street next to the Wall Street bull. And I was like, "Oh, well, that's funny. Like, we should put a claw in front of our um conference." And that's exactly And so so, I asked Devin to research, "Where can I get a lobster in London? Devin comes back with phone numbers and email addresses and websites and I just click through and and think about it and ask it to do some more research and I'll post this guy. That's literally the the lobster that you had was bought from Devin and I think this kind of personal automation for everything else it just matters that you have an agent that has web access that has some smart enough model. I mean this is effectively a claw right here an open claw nano claw whatever the whatever cracker you call it. It doesn't really matter it matters that you're using agents for things that you would otherwise have spent knowledge work on. I might have had executive assistant I might have had a junior employee do these things for me but now I can do it serverless on demand with according agent. I'm not here to only show Devin. Just advise with the company now but I you know I started exploring town because I think what's what's happening here is coding agents going to break in containments right? There's all these other more fit for purpose knowledge management tools like the wikis that Andre Karpathy is is talking about that nano claw that open claw is is is now adopting as well. You're going to see an explosion of this this year. This is like probably the top trend of maybe top three to five trends of 2026 that I want to alert you to. So here is me managing the world's fair in 2020 in in in this summer. Here all the tracks I'm planning. Here's my Apple notes. On the left is my Apple notes of all the people I it's intentionally small and I threw it into into town and on up pops a nicely formatted notion doc with research on all the speakers that I intend to solicit and think about curating. And then obviously once you get enough psychosis you are thinking about replacing entire pieces of SAS. Here is me arguing with my employees about kicking out a SAS tool and building it ourselves because we can. So I clearly have the most psychosis. If anyone of the annoying things is if you're in a position of power or management to deal with employees who are who are not as much in psychosis and try to bring them along the journey and then not talk down or or or or ignore their concerns right? Because they are very valid concerns because they are exactly the people that will have to deal with your when you get it wrong and we do get it wrong. So one one one top one method I I'm approaching this sort of AI replacing SAS concept which I think it it should be relevant for a lot of you is well let's identify the top three concerns and let's systematically reduce reduce them and that's the process that we're going through right now. So I just wanted to give you a little bit of that taste of like here's how AI is changing our business as managing the conference. Um It's come really it's come really a long way. It's a it's a consistent theme I'm seeing even among our our speakers. This is Malte's opening keynote talking about how the the 60% of the the sort of user base of Vercel now is bots is is is agents it's not humans. So actually your dashboards don't matter your APIs matter your CLIs matter your MCPs matter. Here's the MCP apps guys Edo and Liad who spoke today on speaking on ETM about how basically the your custom UI is kind of going away like you should ship UI to somebody else's app and I think like this patterns of like how your primary user is changing is really shifting towards what people are calling agent experience and I think that's something that again I'm really inspired by and focused on because it is helping me right? I no longer care about the Figma dashboard I throw it into cloud co-work and I hope that it works for me. So that's my message agents for everything else are coming. Wake up. Use it. Bring it home to work. If people are insufficiently bought on prescribe them one of these. Thank you. \>\> \[music\]