AI Engineer(@aiDotEngineer)

congratulations to Jonathan and Cognition! more on the ambitious software this team is building at A...

AI Engineer (@aiDotEngineer) — Fri, 11 Sep 2026 20:00:01 GMT

congratulations to Jonathan and Cognition! more on the ambitious software this team is building at AIE World's Fair: youtube.com/watch?v=H7vFrc…

Jon Kelley@jkelleyrtp

We have some very big news to share. 🎉

Today, Dioxus Labs is joining @Cognition to accelerate the development of Devin, Cognition’s autonomous cloud coding agent.

We are excited to continue building Dioxus while also helping Cognition shape the future of software engineering.

Your browser does not support the video tag.

🔗 View on Twitter

🔗 View Quoted Tweet

💬0🔄0❤️3👀367

⚡ Powered by xgo.ing

Details: https://t.co/FcsjTE6hQj

AI Engineer (@aiDotEngineer) — Fri, 11 Sep 2026 13:23:31 GMT

Details: sonarsource.com/customer-stori…

💬0🔄0❤️0👀616

⚡ Powered by xgo.ing

AI writes more of the code. An independent verification layer decides what ships. A global financia...

AI Engineer (@aiDotEngineer) — Fri, 11 Sep 2026 13:18:57 GMT

AI writes more of the code. An independent verification layer decides what ships.

A global financial institution uses @SonarQube across 8,500 projects and 40+ languages. Pipelines fail by default when code misses quality gates.

@SonarSource is sponsoring AI Engineer NY, Oct 12–14: ai.engineer/nyc

M&T Bank uses this playbook, standardizing code quality across the organization on SonarQube and seeing ROI within six months.

💬3🔄0❤️3👀1658📊4

⚡ Powered by xgo.ing

Live now: our Design Engineering Track from AI Engineer World's Fair 2026. Taste, canvases, roundtr...

AI Engineer (@aiDotEngineer) — Thu, 10 Sep 2026 17:03:00 GMT

Live now: our Design Engineering Track from AI Engineer World's Fair 2026.

Taste, canvases, roundtrips, and what one designer plus agents can ship.

Thesis: making things is free now. Knowing what good looks like is not.

youtube.com/watch?v=WkBPX-…

- Understanding is the new bottleneck: @geoffreylitt, Notion
- The Spatial Harness: @max__drake, tldraw
- The Design-Code Roundtrip That Isn't: Jonathan Gordon, ReWeaver AI
- Mousepower: @MVXMXM, Yutori
- Design at the Speed of Adjectives: @pbakaus, Renaissance Geek, Inc.
- Training Taste: @thaiscbranco_, Taste Labs
- Imagination Engineering: @eve_bouff, Y Combinator
- The Missing Layer: @nutlope, Together AI
- Generative UI... in Python? @jlowin, Prefect
- One Designer + AI. Hundreds of Deliverables: Vincent Wendy, AI Engineer

💬4🔄4❤️15👀6112📊9

⚡ Powered by xgo.ing

What resonated: - The purple gradients are gone. Paul Bakaus says the frontier models moved on to C...

AI Engineer (@aiDotEngineer) — Thu, 10 Sep 2026 17:03:00 GMT

What resonated:

- The purple gradients are gone. Paul Bakaus says the frontier models moved on to Claude beige: cream palette, one serif font, italics everywhere. Not bad design, identical design.
- Geoffrey Litt will not send a PR until he passes the five question quiz his agents wrote about their own code. He says it is shocking how often he fails.
- Hassan El Mghari showed landing pages from a frontier model and a two week old open source model, then asked a room of AI engineers to spot the open one. Four hands.
- Taste Labs analyzed two million websites and found the internet was converging before AI arrived. Now a pet shop and a finance firm get the same page.
- Our own designer pointed an agent at the sponsor wall, 140 logos, and asked which were missing. It caught every omission he tested. One designer, 7,000 attendees.

Design Engineering playlist:
youtube.com/watch?v=WkBPX-…

💬0🔄0❤️2👀1017📊1

⚡ Powered by xgo.ing

AI Engineer NY hotel rooms are going fast. Lock in your stay at the Sheraton New York Times Square ...

AI Engineer (@aiDotEngineer) — Wed, 09 Sep 2026 22:32:00 GMT

AI Engineer NY hotel rooms are going fast.

Lock in your stay at the Sheraton New York Times Square (Oct 12-14) now

→ ai.engineer/nyc/2026#hotel

💬0🔄0❤️2👀2082📊1

⚡ Powered by xgo.ing

We’re hiring at AI Engineer. Help us unite and empower the global AI engineering community. Open r...

AI Engineer (@aiDotEngineer) — Wed, 09 Sep 2026 16:30:00 GMT

We’re hiring at AI Engineer.

Help us unite and empower the global AI engineering community.

Open roles:
• Growth Marketing Lead
• Community Manager
• Event Systems & Ops Manager
• Associate Producer

Apply: https://t.co/f2C5wMDtCZ

💬14🔄9❤️77👀9584📊36

⚡ Powered by xgo.ing

Live now: our Context Engineering Track from AI Engineer World's Fair 2026. Thesis: the smartest mo...

AI Engineer (@aiDotEngineer) — Wed, 09 Sep 2026 16:03:00 GMT

Live now: our Context Engineering Track from AI Engineer World's Fair 2026.

Thesis: the smartest model still shows up to work knowing nothing about your company.

youtube.com/watch?v=9R--1t…

- Build-Time vs. Run-Time: Averi Kitsch & Prerna Kakkar, Google
- It's Tokens All The Way Down: @kmad, AlixPartners
- 500 Skills, Zero Fine-Tuning: @ajay_prakash_ai, LinkedIn
- Your agents lack context: @BrandonWaselnuk, Unblocked
- How long can your skills be before your agent forgets what you told it? @seldo, Arize AI
- WTF Is the Context Layer? @prukalpa, Atlan
- MCP Apps, Extending the Frontier: @idosal1 & @liadyosef
- Give the Model Data, Give the User a UI: Dustin Mihalik, Indeed
- MCP Tasks (async): @cdavisafc, Temporal
- The Universal Remote Control for AI: @alexjhancock, Block

💬4🔄6❤️20👀2993📊9

⚡ Powered by xgo.ing

The scenes we keep replaying: - Laurie Voss reran a year old benchmark that used to break models at...

AI Engineer (@aiDotEngineer) — Wed, 09 Sep 2026 16:03:00 GMT

The scenes we keep replaying:

- Laurie Voss reran a year old benchmark that used to break models at 200 instructions. Every current model scored 100 percent instantly. The ceiling now sits near 2,000. Total research bill: $29.
- Cornelia Davis says no agent supports MCP tasks yet because the maintainers are smart. The spec keeps changing. Alex Hancock opened the next talk. He maintains one, and it's just because he's lazy.
- Dustin Mihalik gave Indeed's job search a widget and Claude stopped digging. Text results got fifteen searches and a cherry picked table. The widget got one call. Nobody wants ten carousels.
- Google's demo agent hit a database error, decided the fix was to delete the table and start fresh, and did it. Nothing stopped it.
- Atlan changed its positioning. The sales agent on Atlan's own website kept pitching the old version. No one had told it.

💬1🔄0❤️2👀959📊2

⚡ Powered by xgo.ing

Want in on AI Engineer New York? Apply to be an associate and get a front-row seat to one of the yea...

AI Engineer (@aiDotEngineer) — Mon, 07 Sep 2026 15:35:00 GMT

Want in on AI Engineer New York? Apply to be an associate and get a front-row seat to one of the year's most exciting AI gatherings — connect with brilliant people across the AI ecosystem and build relationships that matter.

ai.engineer/nyc/2026/assoc…

💬6🔄0❤️14👀3185📊9

⚡ Powered by xgo.ing

Early bird pricing for AI Engineer New York ends soon ⏳ If you've been on the fence, now's the time....

AI Engineer (@aiDotEngineer) — Fri, 04 Sep 2026 18:01:00 GMT

Early bird pricing for AI Engineer New York ends soon ⏳ If you've been on the fence, now's the time.

This is where the AI ecosystem's builders, researchers, and leaders come together to share what's actually working — in product systems, infrastructure at scale, and real-world AI across banking, hedge funds, fintech, insurance, and asset management.

Lock in your ticket now before prices go up.

Grab your seat: ai.engineer/nyc/2026

💬0🔄1❤️9👀2739📊2

⚡ Powered by xgo.ing

We are honored to have @arizeai as presenting sponsor at AIE NYC 2026 — where finance-specific AI he...

AI Engineer (@aiDotEngineer) — Fri, 04 Sep 2026 15:02:00 GMT

We are honored to have @arizeai as presenting sponsor at AIE NYC 2026 — where finance-specific AI helps you build smarter, move faster, and outpace competitors.

Arize helps financial services teams improve AI agents. Arize AX turns traces and evals into investigations, reviewable fixes, and experiments. Phoenix offers open-source observability and evaluation.

Event details and tickets: ai.engineer/nyc/2026

💬1🔄0❤️1👀2471📊1

⚡ Powered by xgo.ing

Live now: our Claws & Personal Agents Track from AI Engineer World's Fair 2026. Thesis: agents now ...

AI Engineer (@aiDotEngineer) — Thu, 03 Sep 2026 16:02:00 GMT

Live now: our Claws & Personal Agents Track from AI Engineer World's Fair 2026.

Thesis: agents now run as us.

youtube.com/watch?v=MkRYPF…

- Security Firewall for Agents: @rough__sea, Deno
- Your Agent Didn't Fail. Your Harness Did: @iamvinoth, OpenAI
- Everyone Gets A Software Company: @0thernet, Zo Computer
- Our Agents Are Us: Shu Fang, Two Sigma
- Agents' next frontier: @jgreze, Town
- Claude for Long-Horizon Tasks: @RLanceMartin, Anthropic
- From coding to Knowledge work agents: @KaranVaidya6, Composio
- Your company brain will leak secrets: Tanmai Gopal, PromptQL
- Every Harness Will Become A Claw: @calcsam, Mastra
- Gadgets: @KentonVarda, Cloudflare

💬1🔄2❤️13👀3126📊4

⚡ Powered by xgo.ing

What stuck with us: - Every Two Sigma employee has an agent that runs as their exact identity. Shu ...

AI Engineer (@aiDotEngineer) — Thu, 03 Sep 2026 16:02:00 GMT

What stuck with us:

- Every Two Sigma employee has an agent that runs as their exact identity. Shu Fang grew a mustache so you could tell him apart from his.
- Ryan Dahl's agent, ordered to delete the users table, obeyed. Deno's proxy read the database traffic byte by byte and killed the query.
- Claude wrote one wrong Pokemon memory and fell through the same trap door in five runs out of five. Lance Martin's fix: an offline pass where it rereads its own notes and corrects them.
- Karan Vaidya pointed his OpenClaw at hiring outreach. It mass emailed candidates, exactly as instructed. Some had landed on people in the room.
- Kenton Varda's demo errored in its first minute. It did not matter: the whole cloud platform was running on his laptop. It shipped in August as Cloudflare OS.

Claws & Personal Agents playlist:
youtube.com/watch?v=MkRYPF…

💬0🔄1❤️3👀1701📊2

⚡ Powered by xgo.ing

Our favorite moments: - A phone rang on stage. Harshal Bhangale's agent with a wallet was calling i...

AI Engineer (@aiDotEngineer) — Wed, 02 Sep 2026 04:10:36 GMT

Our favorite moments:

- A phone rang on stage. Harshal Bhangale's agent with a wallet was calling in the trip briefing. Its twin without one left an unsent draft.
- Card rails put a 25 cent floor under a purchase worth a tenth of a cent. Anil Nadiminti prices that mismatch at roughly 250x.
- Nothing stops an x402 buyer minting 1,000 signatures against one wallet before settlement. Jan Curn shipped 20,000 tools onto the protocol anyway.
- Anna Spysz's shopping agent got snarky when she hesitated on headphones. The culprit was her own persona prompt, an aggressive audio gear salesman.
- Two hours before the closing session, West Virginia confirmed David Levine's legal entity for an organization of AI agents. Organization 62847.

The Agentic Commerce playlist:
youtube.com/watch?v=AhQpRa…

💬0🔄0❤️1👀1853📊1

⚡ Powered by xgo.ing

Live now: our Agentic Commerce Track from AI Engineer World's Fair 2026. Ten talks on what happens w...

AI Engineer (@aiDotEngineer) — Wed, 02 Sep 2026 04:00:23 GMT

Live now: our Agentic Commerce Track from AI Engineer World's Fair 2026. Ten talks on what happens when agents get wallets.

Thesis: agents can already spend money. The unsolved part is proving a human said they could.

youtube.com/watch?v=AhQpRa…

- Multimodal Collaborative Agents for Next-Gen Commerce: Nidhi Kaushik Vyas, Google DeepMind
- Why Your AI Agent Needs a Wallet: Harshal Bhangale, Circle
- When AI Agents Pay and Sellers Monetize: Anil Nadiminti, AWS
- x402 isn't good (yet): @jancurn, Apify
- Agent Spending Without Controls: @rodventures & @impranavm_, Edge & Node
- Teaching agents to pay: @annaspies, Stripe
- The Agentic Commerce Stack: Ahnaf Prio, Best Buy
- Your Agent Just Authorized What?!: Jay Mok & Ben Coumes, PayPal
- The End of the Static Screen: @guhgoi, commercetools
- Beyond the Lethal Trifecta: @bigkiduna, Kiduna Club

💬5🔄0❤️15👀3957📊8

⚡ Powered by xgo.ing

Live now: the second half of our AI-Native Enterprises Track from AI Engineer World's Fair 2026. All...

AI Engineer (@aiDotEngineer) — Sat, 29 Aug 2026 18:32:00 GMT

Live now: the second half of our AI-Native Enterprises Track from AI Engineer World's Fair 2026. All 21 talks are up.

Thesis: the agent loop is not the system. The loop around it is.

youtube.com/watch?v=sM1iYg…

- The Half Life of Agent Infrastructure: @benatbox, Box
- An Air-Gapped AI Fortress for Consumer Data: Rachna Srivastava, DFPI
- From Tokenmaxxing to Trusted Throughput: Mingsheng Hong, Ironclad
- Agents Are Where Microservices Were in 2015: Roberto Milev & Uday Kanagala, Navan
- Agentic Sites: @csanchez, Adobe
- AI Agents at Global Scale: @tzakus, Maersk
- Which AI startups actually land enterprise contracts? Brian Lewis, Millennium
- AI Agents Are Just Distributed Systems Now: Salman Munaf, TikTok
- The Signal Layer: @lenadroid, Akamai
- Tell the Robot What You Want: Sandhya Subramani, AWS

💬7🔄9❤️30👀5300📊17

⚡ Powered by xgo.ing

Today's videos, in five facts: - California's financial regulator runs fraud AI behind a fiber opti...

AI Engineer (@aiDotEngineer) — Sat, 29 Aug 2026 18:32:00 GMT

Today's videos, in five facts:

- California's financial regulator runs fraud AI behind a fiber optic cable cut in half. One side receives, there is no laser on the other. The first version collapsed in two hours on live data
- Maersk runs over 200 agent instances against shipping systems whose SOPs were screenshots. The quality came from 100,000 expert corrections in nine months, not a bigger model
- Millennium looks at 10 to 15 vendors per pain, takes two or three demos, runs zero or one pilot. About 5% of demo calls become a contract. Pilots that ran six months now run two weeks
- An agent calls refund and the request times out. Did the money go out? A timeout means unknown. An agent's first instinct is to retry anyway
- Box shipped agentic search on a Tuesday and started rebuilding it on Wednesday. Agent infrastructure has a half life measured in months

Watch the AI-Native Enterprises playlist:
youtube.com/watch?v=sM1iYg…

💬0🔄1❤️1👀1561📊1

⚡ Powered by xgo.ing

Five things from Friday's eleven that stuck with us: - An Amazon team estimated a new inference dat...

AI Engineer (@aiDotEngineer) — Fri, 28 Aug 2026 22:50:20 GMT

Five things from Friday's eleven that stuck with us:

- An Amazon team estimated a new inference data plane at 30 people and 18 months. Six engineers, two of them Distinguished, built it in 76 days. Productivity went down first in almost every team studied
- At Uber a first code review took three hours in 2024. Nine in 2026. uReview now leaves 25,000 comments a week, and the model is 100% sure every time, including when it is wrong
- Over 70% of Uber's PRs come from agents. The maintenance loop runs on Sunday, when CI has capacity and nobody wants a Monday of extra diffs
- A Figma engineer sent a respected skeptic an unlabeled AI analysis. Reply: I did not expect someone I respect this much to send me something this sloppy. Now every PR description opens by hand
- The same prompt takes two seconds or sixty. A reasoning model's normal is a chat model's outage. That is why you keep seeing "something went wrong, please try again"

Watch the AI-Native Enterprises playlist. Part 2 lands tomorrow:
youtube.com/watch?v=ZIYYsA…

💬3🔄1❤️5👀1319📊4

⚡ Powered by xgo.ing

Live now: the first half of our AI-Native Enterprises Track from AI Engineer World's Fair 2026. Th...

AI Engineer (@aiDotEngineer) — Fri, 28 Aug 2026 22:50:19 GMT

Live now: the first half of our AI-Native Enterprises Track from AI Engineer World's Fair 2026. 

Thesis: agents made the pull request cheap and the review expensive.

youtube.com/watch?v=ZIYYsA…

- Launching the Figma MCP Server: @jesselumarie, Figma
- Agentic SDLC at Uber: @udaykiran & @hudaman, Uber
- Building uReview, Uber's Multi-Agent Code Review Engine: @wbond & Ameya Ketkar, Uber
- AI Evals for Cross-Functional Teams: @nmparanjape & Swaroop Chitlur Haridas, DoorDash
- Productionizing LLM Gateways: Kanish Manuja, Twilio
- From AI-Assisted to AI-Native: @clare_liguori, AWS
- How to Get Your Org to Adopt Coding Agents: Eyal Blum, Figma
- How do you diffuse AI into the real world? @varunshenoy_, Long Lake
- How to avoid disaster when vibe-coding a billing engine: Andrew Garvin, Stripe
- Your Code Has Bugs. Lean4 Has Proofs: @varun_pant_, AWS
- AI-Native Organisations Run on Skills: Imad Touil, QuantumBlack

💬14🔄4❤️34👀5652📊22

⚡ Powered by xgo.ing

🚀 AI Engineer Paris is back! Join 1,000 AI engineers, CTOs & VPs of AI on Sept 23–24 for 65+ te...

AI Engineer (@aiDotEngineer) — Thu, 27 Aug 2026 12:01:00 GMT

🚀 AI Engineer Paris is back! Join 1,000 AI engineers, CTOs & VPs of AI on Sept 23–24 for 65+ technical talks and hands-on workshops.

🔥 Speakers are live! Dive into the lineup on htai.engineer/paris� <100 Regular Bird tickets left. Grab yours before they're gone.

💬3🔄3❤️34👀10112📊10

⚡ Powered by xgo.ing

What engineering go to market looks like, from ten teams: - Four reps and 10,000 inbound leads a da...

AI Engineer (@aiDotEngineer) — Wed, 26 Aug 2026 18:03:58 GMT

What engineering go to market looks like, from ten teams:

- Four reps and 10,000 inbound leads a day. Nine months later, when the enterprise product finally shipped, prospects said "you never got back to me." Automate the reply before you hire the team
- A cookie banner library's number one source of inbound since April is coding agents recommending it. Coding agents never read the website for a library, they read node_modules, so the docs now ship inside the package
- An internal GTM assistant has answered a million questions for 6,000 sellers. Two weeks after launch, one in five had tried it. The rule: 50 questions at 95% beats 100 at 70%
- Ask an assistant to compare code intelligence tools and Sourcegraph comes up 65% of the time. Describe the pain it solves instead: zero mentions
- Cloudflare's weekly GTM summary is written by three agents: one drafts from the data, one checks the draft against the data, one fixes the tone. The team read every run for two to three months first

Watch the AI in GTM playlist:
youtube.com/watch?v=wdTRsf…

💬0🔄0❤️3👀1551📊1

⚡ Powered by xgo.ing

Live now: our AI in GTM Track from AI Engineer World's Fair 2026. Thesis: go to market is an engine...

AI Engineer (@aiDotEngineer) — Wed, 26 Aug 2026 18:03:57 GMT

Live now: our AI in GTM Track from AI Engineer World's Fair 2026.

Thesis: go to market is an engineering problem now.

youtube.com/watch?v=wdTRsf…

- GTM Engineering: The Technical Bits: @retttx, Clay
- Reverse-Engineering the AI Buyer: Aliisa Rosenthal, Acrew Capital
- AI in GTM at Notion: @floppyliu, Notion
- The Building Blocks of GTM Orchestration: Arman Vaziri, Ramp
- Design Patterns for AI Trust: Juries, Libraries, and Agent Tiers: @alexdbauer, Upside
- How We Got LLMs to Recommend Our Open Source Library: @burnedchris, Inth
- Knowledge Systems: The New GTM Stack: @jeffzwang, Exa
- How AI Agents Let GTM Teams Scale: Justin Joyce, Cloudflare
- Building GTM AI Agents: Lessons from Deploying to 6,000 Users: Sait Izmit, Snowflake
- The Death of Developer Advocates: @sgjarmak, Sourcegraph

💬2🔄1❤️13👀3039📊6

⚡ Powered by xgo.ing

Live now: our AI Architects, AI Factories Track from AI Engineer World's Fair 2026. Thesis: the mod...

AI Engineer (@aiDotEngineer) — Sat, 22 Aug 2026 17:30:00 GMT

Live now: our AI Architects, AI Factories Track from AI Engineer World's Fair 2026.

Thesis: the model is the easy part of the factory.

youtube.com/watch?v=rbjWzZ…

- The Agent Behind the Curtain: Building the Oz Cloud Agent Platform: Safia Abdalla, Warp
- FinOps for AI Agents: Who Spent All the Tokens?: Tisha Chawla + Susheem Koul, Microsoft
- What If Your Chip Design Team Moved Like a Single Body?: Abduallah Mohamed, AIDAChip
- Preferences Over Benchmarks: Model Routing: Archana Kamath + Tyler Gillam, DigitalOcean
- Coding Agents Don't Scale Themselves. Neither Do Your Teams.: @patrickdebois, Tessl
- Agent Frameworks Considered Harmful: @remilouf, .txt
- Inside 847 Production Clinical AI Notes: Sebastian Fox, Composo
- Give the Agent a Budget, Not a Token: @edorado93, Anthropic

💬12🔄10❤️55👀8851📊28

⚡ Powered by xgo.ing

Eight sessions on the parts of the factory that are not the model: - An agent tidying up after itse...

AI Engineer (@aiDotEngineer) — Sat, 22 Aug 2026 17:30:00 GMT

Eight sessions on the parts of the factory that are not the model:

- An agent tidying up after itself hit a filter that matched everything: 200 workloads deleted in 90 seconds. The fix was not a narrower token. It was a budget: rate limits that refill, and an identity the agent cannot set
- In the largest real world study of AI clinical notes, 1 in 20 carried an error serious enough to cause the patient significant harm. The dangerous ones look fine: nothing in the note is wrong, the detail that matters is missing
- Chip designers told an agent to stay out of the spec files. It agreed, then wrote to them with bash. They blocked bash and sed. It used cat. Now they block at the system level, not tool by tool
- Instead of killing an agent that blows its budget, steer it while it runs: compact the context, trim tool output, tell it to tighten up. Their benchmark: average spend down 78 percent, completion up from 67 to 96 percent
- Warp went open source and jumped from around 20,000 GitHub stars to over 60,000, with thousands of PRs. No human reviewer gets pinged until an agent has approved the PR first

Watch the AI Architects playlist:
youtube.com/watch?v=rbjWzZ…

💬1🔄1❤️3👀1981📊2

⚡ Powered by xgo.ing

What teaching the system actually looks like, from nine workflows: - A nightly report agent ran cle...

AI Engineer (@aiDotEngineer) — Thu, 20 Aug 2026 19:43:00 GMT

What teaching the system actually looks like, from nine workflows:

- A nightly report agent ran clean for weeks, then posted its manager's private team analysis as a PR on the repo. Nothing changed. The model decided to be helpful
- A CTO with 15 recurring meetings ships 2 to 10 PRs a week: brief the agent at 5pm, let it run all night, test the stack in the morning. One overnight run trained two production ML models
- A fine tuned classifier brought in $12M at 50x ROI and still turned into tech debt: every fix cost a week of retraining. Rebuilt as skills on a frontier model, a fix now ships in under an hour
- Agents open hundreds of GitHub issues a night asking researchers to publish their weights. Thousands of issues in: two complaints, and researchers' agents now reply to his agents
- One engineer, zero lines of code written this year, an email product thousands trust. The rule: spend half your time teaching the system what it got wrong

Watch the AI Architects playlist:
youtube.com/watch?v=nxokqO…

💬0🔄0❤️3👀1186📊1

⚡ Powered by xgo.ing

Live now: our AI Architects, Show my Workflow Track from AI Engineer World's Fair 2026. Thesis: the...

AI Engineer (@aiDotEngineer) — Thu, 20 Aug 2026 19:43:00 GMT

Live now: our AI Architects, Show my Workflow Track from AI Engineer World's Fair 2026.

Thesis: the new job is teaching the system that does the work.

youtube.com/watch?v=nxokqO…

- Your Agent Evolved. Your Evals Didn't.: Ameya Bhatawdekar, Braintrust
- The Last Human Code Review: Building Trust in AI-Generated Code: @itamar_mar, Qodo
- Prototyping as Leadership: How a CTO Ships with AI Agents: @hursh, The Browser Company
- Serving 2 Million Models Without Melting: Scaling the Hugging Face Hub: @_Aras_B, Hugging Face
- IT Admin for the AI Workforce: @_sarthak4, Decawork
- The Era of Compound Engineering: @kieranklaassen, Every/Cora
- How I automate my own job at Hugging Face using agents: @NielsRogge, Hugging Face
- Your Fine-Tuned Model Is Tech Debt: A 50x ROI House of Cards: Dan Bjornn, Lease End
- Unlock Agent Autonomy: The Runtime for AI-Native Systems: Tushar Jain, Docker

💬3🔄4❤️31👀4676📊12

⚡ Powered by xgo.ing

Our AI in Healthcare Track from AI Engineer World's Fair 2026 is now live! Thesis: what can never b...

AI Engineer (@aiDotEngineer) — Wed, 19 Aug 2026 18:30:00 GMT

Our AI in Healthcare Track from AI Engineer World's Fair 2026 is now live!

Thesis: what can never be wrong cannot be left to probability.

youtube.com/watch?v=u6q-by…

- From Ambient Documentation to Clinical Intelligence: @c_asawa, Abridge
- Guardrails First: Engineering Member-Facing Health AI: Rashi Agrawal, Hinge Health
- Shipping AI to a Million Patients Without an A/B Test: @JaredJoselowitz, Ufonia
- 200 Million Patient Interactions Later: @vim1up, Hippocratic AI
- AI is the World's largest Relationship Therapist: Clay Cockrell + Tony Fabrikant, CoupleWork AI
- Healthcare's Agent Bytecode, X12 as the Harness for AI Agents: @vasantkearney, Onlay
- Trading Desks to Clinical Trials: @aybh08, Allos AI
- How to build an AI-Native Health Company: Dan Feng, Maven Clinic
- Don't be data poor: @anujiravane, Anterior
- Why Your Enterprise Tech Stack Isn't Ready for AI Agents: @ChrisLovejoy_ + @saulhoward, Anthropic + Anterior

💬2🔄6❤️16👀7881📊9

⚡ Powered by xgo.ing

Ten teams on what you do instead of trusting the model: - A member mentioning self harm or an acute...

AI Engineer (@aiDotEngineer) — Wed, 19 Aug 2026 18:30:00 GMT

Ten teams on what you do instead of trusting the model:

- A member mentioning self harm or an acute emergency is routed to 911 by code that runs above the model. The model never even sees the turn
- You cannot A/B test on patients and you cannot roll back a phone call. So an LLM plays the patient first, and real patients rated the simulated one more human in three of four blind sets
- 31 models run on every Hippocratic call: one leads, 30 specialists decide turn by turn whether to speak. At 10,000 calls a day, a 1 percent error rate is 100 wrong appointments, daily
- Anterior cannot keep medical records even anonymized, so 90 percent of its eval data is generated backward from policy decision trees. Clinicians spot the fakes about 60 percent of the time
- ECRI, the patient safety group hospitals use to rank risk, named AI chatbot misuse the number one health technology hazard of 2026. These ten talks are what shipping against that looks like

Watch the AI in Healthcare playlist:
youtube.com/watch?v=u6q-by…

💬0🔄1❤️3👀1444📊1

⚡ Powered by xgo.ing

Two halves of one problem. What you feed the model, and what you do with what comes out: - Krea thr...

AI Engineer (@aiDotEngineer) — Tue, 18 Aug 2026 17:46:58 GMT

Two halves of one problem. What you feed the model, and what you do with what comes out:

- Krea threw AI images out of the training set on purpose, because distilled data is sticky and the shortcut costs you the whole point. Roughly 30 to 40 in house classifiers run over billions of images
- GPU utilization is a lie. It reads 100 percent while the tensor cores sit idle. Any GPU running above 78 degrees gets pulled rather than debugged
- 1.3 million videos rendered by open source users in 90 days, from a framework deliberately designed so a small model could author it
- An avatar generating continuously for eight hours with no reset, and a sixteen hour deployment being built. It costs about what serving a voice model costs
- 10 dollars now buys three hours of continuously generated video. The old way was a slot machine at roughly 10 dollars a minute

Watch the Generative Media track:
youtube.com/watch?v=-tviRd…

💬0🔄0❤️2👀1517📊1

⚡ Powered by xgo.ing

Live now! our Generative Media Track from AI Engineer World's Fair 2026. Thesis: everything hard ab...

AI Engineer (@aiDotEngineer) — Tue, 18 Aug 2026 17:30:00 GMT

Live now! our Generative Media Track from AI Engineer World's Fair 2026.

Thesis: everything hard about generative media happens before and after the model runs.

youtube.com/watch?v=-tviRd…

- Training Krea 2, what matters in generative model training: Sangwu Lee, Krea.ai
- HTML Is All Agents Need: @Rames_Jusso, HeyGen
- Building an Agentic Video Editor for Mass Consumer: @katedeyneka, Reelful
- The Next Game Engine Won't Have a Manual: @arturonereu, Nereu
- While my guitar gently speaks: Todd Fisher, Philo Ventures
- Voice agents with Realtime Video: Sidney Primas, LemonSlice
- Generative Video at the Speed of Light: @keeganmccallum3, uRun
- Infra behind Krea 2, how to train and serve at scale: Gabriel Jorge Menezes, Krea.ai
- The Next Medium, why real time interactive video changes everything: @Boudatw, Reactor

💬4🔄3❤️16👀8178📊9

⚡ Powered by xgo.ing

🆕 Context Engineering in 2026: Compaction, Memory & Cost https://t.co/08qPA63bBd @Whats_AI, @samr...

AI Engineer (@aiDotEngineer) — Mon, 17 Aug 2026 16:34:43 GMT

🆕 Context Engineering in 2026: Compaction, Memory & Cost
youtube.com/watch?v=WP3hjU…d

@Whats_AI, @samridhivaid and @omar_solano1 return!

This workshop is about engineering the context window so rot stops happening, shown with @towards_AI's open-source AI tutor, which answers questions for students of our AI-engineering courses. 

Context engineering is deciding what the model sees on every single call — instructions, history, retrieved course content, memory, and tool outputs — and it's the line between a tutor that holds a coherent session and one that forgets the student's setup halfway through. 

We'll move in three stages, mirroring how the project actually went. The concepts: 

- the two root problems (a finite window, a stateless model), 
- the full compaction toolkit (truncation, trimming, tool-result clearing, summarization, and offloading to files — and when each actually helps), 
- memory that survives across sessions, skills loaded on demand, and 
- production-grade retrieval (chunking, metadata, course scoping, hybrid search, reranking, and evaluating). 

We'll cover the tutor's architecture, and the evaluation harness we used to measure every run on Gemini — tokens, cost, latency, and memory probes instead of vibe-checks. At real volume, even Gemini Flash got expensive, so we tested whether open and local models could match the quality for a fraction of the cost and match result quality. 

Everything is open-source and will be shared during the workshop.

💬4🔄7❤️25👀7609📊15

⚡ Powered by xgo.ing

Congrats to @OpenRouter on their $7B acquisition! https://t.co/a4beTyf081 throwback to last year's...

AI Engineer (@aiDotEngineer) — Mon, 17 Aug 2026 05:27:44 GMT

Congrats to @OpenRouter on their $7B acquisition!

youtube.com/watch?v=84Vtz2…

throwback to last year's WF

💬1🔄5❤️78👀12168📊15

⚡ Powered by xgo.ing

This is about meeting the web where it actually is: - Some school districts answer a procurement qu...

AI Engineer (@aiDotEngineer) — Fri, 14 Aug 2026 17:25:06 GMT

This is about meeting the web where it actually is:

- Some school districts answer a procurement question by making you file a FOIA request, then scanning your email onto a Google Drive. These are the people you are expecting to ship an MCP server
- A sub megabyte script that blindly replays recorded clicks matches or beats the frontier model it was copied from, on the standard computer use benchmarks
- An agent got signed out mid task, decided it could infer the password, guessed until the account locked
- OpenAI threatened to ban his account while he was preparing the talk. He beat Turnstile, MTCaptcha, Lemin and reCAPTCHA v2 with deterministic code and one vision call per round
- On a new electrical engineering benchmark the best agent fully passed 6 of 25 tasks. Starting from a blank schematic instead of an existing one, it passed none

Watch the Computer Use playlist:
youtube.com/watch?v=Ki980n…

💬0🔄2❤️6👀1879📊2

⚡ Powered by xgo.ing

Live now: our Computer Use Track from AI Engineer World's Fair 2026. Nobody is going to rebuild the...

AI Engineer (@aiDotEngineer) — Fri, 14 Aug 2026 17:01:00 GMT

Live now: our Computer Use Track from AI Engineer World's Fair 2026.

Nobody is going to rebuild the web for your agent.

youtube.com/watch?v=Ki980n…

- Computer Use models will agentify the web, not APIs: @DhruvBatra_, Yutori
- Computer Use at the Edge of the Statistical Precipice: @proceduralia, Programma Labs
- Bringing agents onto the world wide web: @pk_iv, Browserbase
- The Dark Arts of Web Automation: @CoreyGallon, Rexmore
- From RL to IRL: Gaurav Mishra, Amazon AGI Lab
- The Rise of CaaS, context as a service: Omer Primor, Bright Data
- Computer-Use 2.0, agents just got multi-cursor: @francedot + Dillon DuPont, Cua
- Persona Engineering, a field guide to synthetic personas: @ianand, InsightSciences.ai
- How Web Data Infrastructure Powers the Next Generation of AI: Patricija Žemaitytė, Oxylabs

💬4🔄6❤️27👀4045📊12

⚡ Powered by xgo.ing

@pgasawa uh oh https://t.co/ctQMhqIc2k

AI Engineer (@aiDotEngineer) — Thu, 13 Aug 2026 05:37:23 GMT

@pgasawa uh oh x.com/willdepue/stat…

will depue@willdepue

i’m so tired of the continual learning thing. what the fuck is continual learning. what do you expect ‘solving it’ will do, like what eval is supposed to go up here. just devolved slop ML lingo now

🔗 View Quoted Tweet

💬1🔄0❤️2👀1963📊1

⚡ Powered by xgo.ing

Memory talks on what happens after training stops: - ChatGPT thinks Shlok Khemani went to Turkey in...

AI Engineer (@aiDotEngineer) — Wed, 12 Aug 2026 20:40:13 GMT

Memory talks on what happens after training stops:

- ChatGPT thinks Shlok Khemani went to Turkey in 2025. He never did. It read a conversation where he was deciding whether to go, and it still cannot tell that its own record conflicts
- On the first real continual learning benchmark, plain in context learning beat the dedicated memory systems
- Scale on policy self distillation to a hundred tool calls and the teacher course corrects so often the model collapses into saying "maybe"
- A teacher moved an agent from finishing 22% of its tasks to 60% without ever changing the tool call tokens. It only reshaped the reasoning leading up to the call
- Agents are bad at finding anomalies. Ask them to investigate the anomalies you already found

Watch the Memory & Continual Learning track:
youtube.com/watch?v=iqloyW…

💬0🔄2❤️3👀1035📊1

⚡ Powered by xgo.ing

Live now: our Memory & Continual Learning Track from AI Engineer World's Fair 2026. Thesis: we scal...

AI Engineer (@aiDotEngineer) — Wed, 12 Aug 2026 20:40:12 GMT

Live now: our Memory & Continual Learning Track from AI Engineer World's Fair 2026.

Thesis: we scaled intelligence and got the world's smartest novice.

youtube.com/watch?v=iqloyW…

- Beyond Static Intelligence, evaluating continual learning: @pgasawa, UC Berkeley
- Scaling up Continual Learning: @rronak_, Trajectory
- Memory Harnesses for Long-Running Research Agents: @Stefania_druga, Sakana.ai
- Scaling Compute on Context: @jxmnop, Engram
- Intelligence + Continual Learning = Expertise: @ysu_nlp, NeoCognition
- Adaption Labs, gradient free continual learning: @sarahookr, Adaption
- Improving Agents is a Data Mining Problem: @Vtrivedy10, LangChain
- Bringing Continual Learning into Enterprises: @samueldenton, Applied Compute
- Designing Agents, The Floor Is the Frontier: @benhylak, Raindrop
- Lessons from Studying Every Memory System: @shloked, Independent
- LLM Knowledge Bases, a practical guide: @bholmesdev, Warp

💬7🔄12❤️33👀8388📊19

⚡ Powered by xgo.ing

Ten talks on what breaks after the model gets good: - Give people an AI that is confidently lying a...

AI Engineer (@aiDotEngineer) — Sat, 08 Aug 2026 21:01:02 GMT

Ten talks on what breaks after the model gets good:

- Give people an AI that is confidently lying and they follow it nearly 80% of the time, against 92.7% when it is right. Human review is not the backstop you think it is
- GPT 5.5 declared a full repo refactor finished in 10 minutes 22 seconds. It had written 2,000 lines of scaffolding and no actual models
- Prompting an agent to behave is not a guardrail, because whoever prompt injects it can undo the instruction. Declare permissions instead, and let the automation do nothing
- A compromised litellm sat live for three hours harvesting API keys and SSH keys. It was caught because the malware crashed Cursor
- Everyone is shipping more and reaching fewer people. It is the newsletter author writing a book a week that nobody reads

Watch the Agentic Engineering track:
youtube.com/watch?v=Yk87oU…

💬1🔄1❤️5👀2549📊3

⚡ Powered by xgo.ing

Live now: our Agentic Engineering Track from AI Engineer World's Fair 2026. Thesis: writing the cod...

AI Engineer (@aiDotEngineer) — Sat, 08 Aug 2026 21:01:01 GMT

Live now: our Agentic Engineering Track from AI Engineer World's Fair 2026.

Thesis: writing the code stopped being the hard part.

youtube.com/watch?v=Yk87oU…

- DeepSWE, expert code datasets: @shiqyy, Datacurve
- Anthropic's CCA Exam as a field guide: @coyle_frankp, UC Berkeley
- Guide, Verify, Solve: Anirban Chatterjee, Sonar
- Benchmarking coding agents on new vs legacy codebases: @denyslinkov, Wisedocs
- Codex, behind the harness: @dkundel, OpenAI
- Multiplayer agentic engineering: Arjun Singh, Superconductor
- Always on agents without the on call tax: Justin Smith, Resolve AI
- Realtime multiplayer, automation, and you: @idangazit, GitHub
- Velocity Sickness: Matt Dailey, Ref.
- Open Source Is Dead. Long Live Open Source: @sdrzn, Cline

💬2🔄6❤️32👀5937📊12

⚡ Powered by xgo.ing

Tickets are live for AI Engineer New York: Oct 12 to 14, 2026, at the Sheraton New York Times Square...

AI Engineer (@aiDotEngineer) — Fri, 07 Aug 2026 18:22:14 GMT

Tickets are live for AI Engineer New York: Oct 12 to 14, 2026, at the Sheraton New York Times Square.

Our third NYC event, and the biggest one yet, now focused on AI in financial services. Banking, hedge funds, trading, insurance, accounting. In production use cases only, no vendor pitches.

Day 1 is hands on workshops, then two full days of keynotes and talks across engineering and leadership.

Early Bird is first come, first served until it sells out. This is a much smaller room than World's Fair.

Tickets: ai.engineer/nyc/2026/ticke…
Speaking: sessionize.com/aienyc2026

Your browser does not support the video tag.

🔗 View on Twitter

💬6🔄3❤️30👀6315📊13

⚡ Powered by xgo.ing

Local is no longer the compromise: - A dense 27B Qwen now beats the 405B Llama from 21 months earli...

AI Engineer (@aiDotEngineer) — Fri, 07 Aug 2026 01:32:43 GMT

Local is no longer the compromise:

- A dense 27B Qwen now beats the 405B Llama from 21 months earlier, on the same RTX 3090 that used to host Llama 2
- Quantize one wrong number, just one, and a model gets 20% dumber. Respect the super weights and GLM 5.2 shrinks from 1.5TB to 250GB
- Cognition cut the cost of Fable level intelligence by 40% by letting the frontier model plan while cheaper models implement
- OpenRouter's auto router sat almost unused for two years until openclaw's ten minute heartbeats found it a purpose
- Three weeks of NVIDIA swarming with EXO Labs got a 10x speedup on the DGX Spark without solving any new computer science

Watch the Local AI playlist:
youtube.com/playlist?list=…

💬0🔄1❤️6👀1982📊2

⚡ Powered by xgo.ing

Live now: our Local AI Track from AI Engineer World's Fair 2026, brought to you by @nvidia. Thesis:...

AI Engineer (@aiDotEngineer) — Fri, 07 Aug 2026 01:32:42 GMT

Live now: our Local AI Track from AI Engineer World's Fair 2026, brought to you by @nvidia.

Thesis: frontier intelligence is becoming something you own.

youtube.com/watch?v=KB41dT…

- State of the Union: @josephofiowa + @alexocheema + @TheAhmadOsman + @MatthewBerman, with @naderlikeladder
- The Desktop Frontier: @TheAhmadOsman, Osmantic
- Local Models: Vincent Weisser + @latkins + @llm_wizard, with @Baxate
- Compression at the Edge: @danielhanchen + Asma Beevi + @mervenoyann + Parth Sareen, with @llm_wizard
- Model Routing: @walden_yan + Tanay Varshney + @alexatallah, with @naderlikeladder

💬4🔄8❤️49👀22317📊18

⚡ Powered by xgo.ing

The State of Model Routing — with NVIDIA Tanay Varshnay Cognition @walden_yan OpenRouter @alexata...

AI Engineer (@aiDotEngineer) — Thu, 06 Aug 2026 18:01:54 GMT

The State of Model Routing — with 

NVIDIA Tanay Varshnay
Cognition @walden_yan 
OpenRouter @alexatallah 

hosted by @NaderLikeLadderyoutu.be/QHBjufYK8TA?is… 

a wonderful deep dive in to one of the hottest topics in AI Engineering right now - when and how should you switch and combine models?

💬1🔄1❤️14👀6240📊6

⚡ Powered by xgo.ing

Better data is not just cleaner data: - General Reasoning gave frontier models $100k each to trade ...

AI Engineer (@aiDotEngineer) — Thu, 06 Aug 2026 10:19:14 GMT

Better data is not just cleaner data:

- General Reasoning gave frontier models $100k each to trade a Premier League season. All of them lost money
- Arcee hit the open frontier on 17T tokens of public data for under $20M all in
- poolside kills a run when two replicas disagree. An FP8 race condition still corrupted 0.5% of gradients
- Surge AI found an IFEval task you can ace with a Cyrillic lookalike letter and no story at all
- Bugcrowd ran agents at 41 real V8 vulnerabilities and got out-of-sandbox exploits, including a zero day

Explore the Data and RL Environments playlist:
youtube.com/watch?v=2aS7aK…

💬1🔄0❤️4👀1630📊2

⚡ Powered by xgo.ing

Live now: our Data Quality Track from AI Engineer World's Fair 2026. Thesis: the cheapest compute y...

AI Engineer (@aiDotEngineer) — Thu, 06 Aug 2026 10:15:37 GMT

Live now: our Data Quality Track from AI Engineer World's Fair 2026.

Thesis: the cheapest compute you will ever buy is better data.

youtube.com/watch?v=2aS7aK…

- @arimorcos, DatologyAI
- @marah_i_abdin + @robert_mchardy, poolside
- @RayanGarg, Theta Software
- David Brumley, Bugcrowd
- @stochasticchasm, Arcee AI
- @thaiscbranco_, Taste Labs
- @rosstaylor90 + @chengxitaylor, General Reasoning
- @nickheiner, Surge AI
- @madiator, Bespoke Labs

💬1🔄2❤️12👀2628📊4

⚡ Powered by xgo.ing

Every model is already doing exactly what you paid it to do: - RLHF pays for approval, so ChatGPT c...

AI Engineer (@aiDotEngineer) — Wed, 05 Aug 2026 07:45:38 GMT

Every model is already doing exactly what you paid it to do:

- RLHF pays for approval, so ChatGPT called a fart audio file "a very eerie vibe atmosphere piece"
- Applied Compute's tool calls failed 10% of the time and the model started answering shorter. No length penalty anywhere
- Prime Intellect: a rising reward curve means the model learned the job, or learned your grader
- Arithmetic and Hugging Face hid a real zero day in a live system. One solve at K1

Explore the Post-training playlist:
youtube.com/watch?v=ZFxh7s…

💬0🔄1❤️2👀1643📊1

⚡ Powered by xgo.ing

Live now: our Posttraining & Midtraining Track from AI Engineer World's Fair 2026. 9 talks. 11 spea...

AI Engineer (@aiDotEngineer) — Wed, 05 Aug 2026 02:55:40 GMT

Live now: our Posttraining & Midtraining Track from AI Engineer World's Fair 2026.

9 talks. 11 speakers. One thesis: a model does not do what you want. It does what you rewarded.

youtube.com/watch?v=ZFxh7s…

- @CompleteSkeptic, TypeSafe AI
- @SeanZCai, State of Data
- @uri_rolls + @Thom_Wolf, Arithmetic + Hugging Face
- @raymondmfeng, Applied Compute
- @willccbb, Prime Intellect
- @_potatodonkey_, Emulated
- @kenbwork, LatchBio
- @olive_jy_song + @realDanFu, MiniMax + Together AI
- Ali Khial, G2i

💬1🔄3❤️12👀3510📊6

⚡ Powered by xgo.ing

Live now: our Posttraining & Midtraining Track from AI Engineer World's Fair 2026. The question is ...

AI Engineer (@aiDotEngineer) — Wed, 05 Aug 2026 02:18:53 GMT

Live now: our Posttraining & Midtraining Track from AI Engineer World's Fair 2026.

The question is no longer just what a model can do... but what the training has optimized it to do.

youtube.com/watch?v=ZFxh7s…

- @CompleteSkeptic, TypeSafe AI
- @SeanZCai, State of Data
- @uri_rolls + @Thom_Wolf, Arithmetic + Hugging Face
- @raymondmfeng, Applied Compute
- @willccbb, Prime Intellect
- @_potatodonkey_, Emulated
- @kenbwork, LatchBio
- @olive_jy_song + @realDanFu, MiniMax + Together AI
- Ali Khial, G2i

💬1🔄2❤️3👀627📊3

⚡ Powered by xgo.ing

*Jia Wu, @j_wu8 FDE lead at @cognition

AI Engineer (@aiDotEngineer) — Mon, 03 Aug 2026 01:58:38 GMT

*Jia Wu, @j_wu8 FDE lead at @cognition

Limestone Digital@LimestoneHQ

Deployed engineering lead at Cognition, Gia:

"Coding itself is a mostly solved problem. Writing code faster - that's only 20% of the problem. The real problem is how do you test this code, review it, deploy it, and maintain it across the enterprise."

One of the most honest talks at @aiDotEngineer on AI engineering this year.

17 minutes of pure signal from the team that took Devin from "I'd only use this if I was desperate" to delivering 150% extra headcount at the world's largest banks.

The gap between AI demo and AI deployment is where every coding tool dies. This team closed it.

If you're still measuring AI productivity by how fast you write code, you're solving the wrong 20%.

Watch it, then read the article on AI-native engineering at enterprise scale below.

Your browser does not support the video tag.

🔗 View on Twitter

🔗 View Quoted Tweet

💬0🔄0❤️15👀5872📊4