Built with Astro, hosted on Cloudflare. Domain registered in 2026.
aietalks.com is built with Astro, hosted on Cloudflare. It measures traffic with Cloudflare Web Analytics and Plausible Analytics.
| Technology | Version | First seen |
|---|---|---|
| JavaScript Frameworks & Libraries1 | ||
| – | 7 Nov 2026 | |
| CDN & Hosting1 | ||
| – | 7 Nov 2026 | |
| Analytics & Tag Management2 | ||
| – | 7 Nov 2026 | |
| Fonts & Icon Fonts1 | ||
| – | 7 Nov 2026 | |
| Site Search1 | ||
| – | 7 Nov 2026 | |
Registration from the registry's RDAP record, the certificate the site serves, and its rank in public traffic lists.
The crew hasn’t read this site’s colors yet. We read a palette when we capture a site’s screenshot, busiest sites first. Meanwhile, browse palettes from other sites.
Counted from every stack change we've recorded: of the sites we watched running one of this site's tools, how many went on to add another. A pattern across sites, not a forecast for this one.
arnold.ns.cloudflare.comhadlee.ns.cloudflare.comgoogle-site-verification=7nu74UitumOlcbFhH3USVMdRzRp6TkAB0mWhlkHR5EgUser-agent: * Allow: / Sitemap: https://aietalks.com/sitemap-index.xml
# AIE Talks > Summaries of talks from the AI Engineer YouTube channel (https://www.youtube.com/@aiDotEngineer). Each talk page has a three-bullet TL;DR, a summary, five to nine key ideas with links into the video at the second they start, quotes, and tags. Curated packs put talks in a deliberate order. Every talk page and pack page also exists as plain Markdown: add .md to any talk or pack URL, or send Accept: text/markdown to the same URL. ## Packs - [Coding agents on real codebases](https://aietalks.com/packs/coding-agents-on-real-codebases): Coding agents look capable in a fresh repository and come unstuck in a ten-year-old one. - [Agents in production: reliability, evals and cost](https://aietalks.com/packs/agents-in-production): A production agent can give the right answer and still refund the same customer twice, act on a stale record or quietly finish half a job. - [Robotics: when models meet the physical world](https://aietalks.com/packs/robotics): A robot policy can understand the scene and still fail because its training data is thin, its controller is late or its motors cannot keep up. - [Harness engineering](https://aietalks.com/packs/harness-engineering): When an agent forgets its plan, edits the wrong file or declares victory without changing anything, the model is only one suspect. - [Beyond the chat box](https://aietalks.com/packs/beyond-the-chat-box): The chat box makes users package their work into prompts, wait, inspect the answer and try again. - [Software factories](https://aietalks.com/packs/software-factories): A software factory can build quickly and still fail if nobody decides what is worth building or retains enough understanding to support it after release. - [Memory and continual learning](https://aietalks.com/packs/memory-and-continual-learning): An agent can remember your name and still repeat yesterday's mistake. - [AI in high-risk domains](https://aietalks.com/packs/high-risk-domains): High-risk AI is not one field, and its controls cannot be copied whole from one industry to another. - [Local AI on your own hardware](https://aietalks.com/packs/local-ai): A personal agent may need your messages, calendar and accounts, which makes sending every thought and action through a cloud provider a hard bargain. - [Data for models and agents](https://aietalks.com/packs/data-for-models-and-agents): Training data is not a pile of tokens waiting to become a model. - [Context engineering](https://aietalks.com/packs/context-engineering): More context can make an agent worse. - [Search and retrieval](https://aietalks.com/packs/search-and-retrieval): A fluent RAG answer can still be wrong because the right passage never reached the model. - [Security for agents](https://aietalks.com/packs/security): An agent can return a reassuring answer after it has already leaked a credential, crossed into another user's data or deleted something real. - [Post-training and RL](https://aietalks.com/packs/post-training-and-rl): In one financial tool-use task, a trained 4 billion parameter model outperforms a 235 billion parameter baseline, but only because the team knows which behavior to reward. - [Voice agents in real conversations](https://aietalks.com/packs/voice): A voice agent can have a strong model and still fail because it talks over someone, leaves a dead pause or hands a caller to a person with no useful context. - [Inference: serving models efficiently](https://aietalks.com/packs/inference): A model can answer correctly and still be unusable once real prompts, concurrent users and a GPU bill arrive. - [MCP beyond the demo](https://aietalks.com/packs/mcp): An MCP server that answers one tool call on your laptop has proved little. - [Agent skills](https://aietalks.com/packs/skills): An API tells an agent what it can call, but not how your team gets a job right. - [Evals that survive production](https://aietalks.com/packs/production-evals): An eval set assembled before launch can tell you whether a change helped on the cases you expected. - [Benchmarks worth trusting](https://aietalks.com/packs/benchmarks): A benchmark can rank models cleanly while rewarding the wrong work. - [Image generation beyond the prompt](https://aietalks.com/packs/image-generation): Image generators are impressive until you need the same character twice, a product to keep its shape, or an edit to work for millions of people. - [Video generation](https://aietalks.com/packs/video-generation): Video generation stops being useful when every clip demands a long wait, another prompt and a manual rescue in the editor. ## Talks - [7 Things Your AI Coworker Needs](https://aietalks.com/talks/7-things-your-ai-coworker-needs): Philipp Schmid, Google DeepMind. Philipp Schmid explains how to build an autonomous finance coworker with Gemini and the Gemini API's Managed Agents. He breaks the system into seven parts:... - [AI Agents Will Upend Investing Even More Than Coding](https://aietalks.com/talks/ai-agents-will-upend-investing-even-more-than-coding): Frank Long, Coatue. Frank Long argues that agents could affect investing more than coding because investing compresses a huge amount of information into a small decision: what to... - [Your AI Agent Keeps Making the Same Mistakes](https://aietalks.com/talks/your-ai-agent-keeps-making-the-same-mistakes): Sparsh Jain, CentralAgent AI. Sparsh Jain argues that an agent with more stored information does not necessarily become better. At enterprise scale, dumping all available context into a prompt... - [Your AI Isn't the Advantage. Your Judgment Is](https://aietalks.com/talks/your-ai-isnt-the-advantage-your-judgment-is): Sid Pednekar. Sid Pednekar argues that access to AI is now widespread, while business results remain uneven. He says the scarce resource is judgment, which means choosing... - [How Netflix Evaluates AI When There's No Right Answer](https://aietalks.com/talks/how-netflix-evaluates-ai-when-theres-no-right-answer): Cameron Wolfe, Netflix. Cameron Wolfe presents a three-part evaluation process for AI tasks where quality cannot be checked with a deterministic function. Human evaluation comes first. Subject-matter experts... - [How We Took an AI Agent From 41% to 97% Accurate](https://aietalks.com/talks/how-we-took-an-ai-agent-from-41-to-97-accurate): Milind Sunil Shah, Brainbase Labs. Milind Sunil Shah explains how Brainbase Labs improved an agent that collected competitor pizza prices for Domino's. The agent had to select the right store,... - [Stop Giving Your AI Agents API Keys](https://aietalks.com/talks/stop-giving-your-ai-agents-api-keys): Kim Maida, Keycard & Mason Egger, Temporal. Kim Maida demonstrates an expense desk agent that starts with shared API keys and no reliable record of who filed or approved an expense. She... - [Stop Making Your Customers Click Approve](https://aietalks.com/talks/stop-making-your-customers-click-approve): Emil Chetty, Wise. Emil Chetty describes a payment workflow in which the customer approves a complete job instead of supervising every agent step. Wise first uses read-only access... - [Why Waiting for the Next AI Model Can Pay Off](https://aietalks.com/talks/why-waiting-for-the-next-ai-model-can-pay-off): John Stecher, Blackstone & Scott Worland, Norm AI. John Stecher and Scott Worland discuss how to build AI systems while models keep changing. Stecher puts data strategy first: companies need to move information... - [AI Labs Will Trap You or Replace You](https://aietalks.com/talks/ai-labs-will-trap-you-or-replace-you): Dillon Rolnick, Nous Research. Dillon Rolnick argues that falling model prices will force AI labs to find other ways to recover the enormous cost of training and serving their... - [Building the AI Agent Is Easy. Proving It's Safe Isn't](https://aietalks.com/talks/building-the-ai-agent-is-easy-proving-its-safe-isnt): Jonathan Osborne, NatWest. Jonathan Osborne describes NatWest's move from scripted answers in Cora to AI-generated answers and agents that can take actions for customers. The agent itself was... - [Don't Approve the AI Model. Approve the App](https://aietalks.com/talks/dont-approve-the-ai-model-approve-the-app): Doneyli De Jesus, Langfuse. Doneyli De Jesus describes work with a top-five bank that needed deployment controls for AI agents. Public model benchmarks were not useful because they did... - [Our Smaller Model Beat Opus on Finance Tasks](https://aietalks.com/talks/our-smaller-model-beat-opus-on-finance-tasks): Alexander Shevchenko, Ramp. Alexander Shevchenko argues that application companies still need applied AI research even as frontier models become cheaper, faster, and more capable. Ramp's teams shipped agents... - [Robots Will Be Trained on Video Games](https://aietalks.com/talks/robots-will-be-trained-on-video-games): Pim de Witte, General Intuition. Pim de Witte argues that models for acting should be built around continuous observation and action. LLMs work well when an interaction unfolds in turns,... - [Why AI Makes You 100x Faster but Not Your Company](https://aietalks.com/talks/why-ai-makes-you-100x-faster-but-not-your-company): Mitchell Troyanovsky, Basis. Mitchell Troyanovsky argues that the large productivity gains engineers get from tools such as Codex and Claude Code do not spread through companies automatically. Organizations... - [Why Datasite Stopped Building Its Own Parsing](https://aietalks.com/talks/why-datasite-stopped-building-its-own-parsing): Adit Abraham, Reducto & Raj Bakhru, Datasite. Raj Bakhru describes how his BlueFlame startup became part of Datasite and brought AI into a platform containing billions of pages from sensitive deal documents.... - [Better Search Took This AI Agent From 36% to 98% Accuracy](https://aietalks.com/talks/better-search-took-this-ai-agent-from-36-to-98-accuracy): Aamir Shakir, Mixedbread. Aamir Shakir argues that many agent failures begin before the language model sees any context. OCR can lose tables and diagrams, chunking can split information... - [Make Your AI Agents Predictable With Graph Engineering](https://aietalks.com/talks/make-your-ai-agents-predictable-with-graph-engineering): Zach Blumenfeld, Neo4j. Zachary Blumenfeld describes graph engineering as designing agent work as an explicit graph. Execution graphs route work at runtime, context graphs store conversations, entities, decisions,... - [Without Caching, Your AI Agent Costs 10x More](https://aietalks.com/talks/without-caching-your-ai-agent-costs-10x-more): George Cameron, Artificial Analysis. George Cameron explains how developers should measure and control the cost of AI agents. Artificial Analysis evaluates models across intelligence, speed, pricing, cache hit rate,... - [You Can't Tell What's Real Anymore. Here's What Comes Next](https://aietalks.com/talks/you-cant-tell-whats-real-anymore-heres-what-comes-next): Cristóbal Valenzuela, Runway. Cristóbal Valenzuela describes Runway's progress from a 128-by-128-pixel AI film in 2019 to video that people cannot reliably distinguish from real footage. He argues that... - [Your AI Agents Shouldn't Care Where They Run](https://aietalks.com/talks/your-ai-agents-shouldnt-care-where-they-run): Nikhil Harithas, Factory. Nikhil Harithas explains how Factory rebuilt Droid for customers whose security and deployment requirements made the usual agent setup unusable. He uses Blackstone as the... - [A Valid Key Let an AI Agent Delete Production](https://aietalks.com/talks/a-valid-key-let-an-ai-agent-delete-production): Sagar Batchu, Speakeasy. Sagar Batchu uses an incident involving a coding agent to explain why valid credentials are not enough for AI systems. The agent was asked to... - [AI Writes Code Faster Than Your Team Can Review It](https://aietalks.com/talks/ai-writes-code-faster-than-your-team-can-review-it): Andrew Norrish, Sourcegraph. Andrew Norrish argues that AI coding tools have increased developer output faster than teams can safely review it. Pull requests are larger and more numerous,... - [I Don't Fear AI Slop. I Fear Untold Stories](https://aietalks.com/talks/i-dont-fear-ai-slop-i-fear-untold-stories): Zack Proser, WorkOS. Zack Proser argues that AI-assisted writing should be judged by the finished work and the human responsibility behind it. He describes Blogbot, a Slack tool... - [Your AI Pilots Won't Move the P&L. Here's What Will](https://aietalks.com/talks/your-ai-pilots-wont-move-the-p-l-heres-what-will): Adam Nahari, Apollo Global Management. Adam Nahari argues that AI transformation is mainly a change-management problem. General-purpose copilots can improve individual productivity, but those gains often get competed away or... - [Your Company Should Train Its Own AI Models](https://aietalks.com/talks/your-company-should-train-its-own-ai-models): Will Brown, Prime Intellect. Will Brown argues that AI-native companies will train models on their own workflows instead of relying only on frontier models and long prompts. Companies contain... - [I Made a Fruit Fly Review My Code](https://aietalks.com/talks/i-made-a-fruit-fly-review-my-code): Nick Nisi, WorkOS. Nick Nisi argues that human code review is changing because code generation and automated checks are increasing the number of pull requests. He built a... - [If Your AI Scores 100%, Your Eval Is Wrong](https://aietalks.com/talks/if-your-ai-scores-100-your-eval-is-wrong): Ramakrishnan Lokanathan, eSlams. Ramakrishnan Lokanathan explains how to evaluate AI systems used in high-stakes applications. He defines an eval as a set of tasks, standards, and evidence used... - [Most of Your Docs Readers Are Now AI Agents](https://aietalks.com/talks/most-of-your-docs-readers-are-now-ai-agents): Hahnbee Lee, Mintlify & Lincoln Murr, Coinbase. Hahnbee Lee and Lincoln Murr discuss what changes when AI agents become major users of software. Lee says Mintlify has seen agent traffic rise from... - [Why We Rebuilt Microsoft Office From Scratch for AI Agents](https://aietalks.com/talks/why-we-rebuilt-microsoft-office-from-scratch-for-ai-agents): Yijia Liang, Compound AI. Yijia Liang explains why personal AI agents do not directly fit institutional finance. A deal team needs several people to share one agent, keep work... - [Everyone Has the Same AI Models. Here's What Sets Agents Apart](https://aietalks.com/talks/everyone-has-the-same-ai-models-heres-what-sets-agents-apart): Richmond Alake, Oracle. Richmond Alake argues that access to the same models and information does not give agents the same intelligence. He uses the World War II example... - [Voice AI Where One Wrong Question Breaks the Rules](https://aietalks.com/talks/voice-ai-where-one-wrong-question-breaks-the-rules): Shiv Indap, Upstart & Brooke Hopkins, Coval. Shiv Indap and Brooke Hopkins describe how Upstart evaluates a voice agent that handles loan verification calls. The agent asks about continuing income, use of... - [When Running Your Own AI Models Beats Paying for APIs](https://aietalks.com/talks/when-running-your-own-ai-models-beats-paying-for-apis): Steven Latré, imec. Steven Latré presents imec's study of the cost of moving long-running coding agents from hosted APIs to owned infrastructure. The team ran more than a... - [Why A/B Testing for AI Products Is Weird](https://aietalks.com/talks/why-a-b-testing-for-ai-products-is-weird): Supriya Subudhi, Bumble. Supriya Subudhi explains why a positive A/B test is only the beginning for generative AI features. Bumble tested AI photo feedback for dating profiles. Although... - [Your AI Agent Wants Your Data. Here's How to Lock It Down](https://aietalks.com/talks/your-ai-agent-wants-your-data-heres-how-to-lock-it-down): Jeff Smith, Oracle. Jeff Smith explains why database access is useful for agents and why it needs controls. Oracle's AI Database stores vectors alongside relational and other data... - [How to Decide What Your AI Agent Should Remember](https://aietalks.com/talks/how-to-decide-what-your-ai-agent-should-remember): Wojciech Pluta, Oracle. Wojciech Pluta explains why agent memory needs deliberate write controls. Retrieval-augmented generation lets an agent read information, but retrieval alone does not decide what should... - [The Cheaper AI Model That Costs You More](https://aietalks.com/talks/the-cheaper-ai-model-that-costs-you-more): Rayan Krishnan, Vals AI. Rayan Krishnan argues that broad model card scores are too coarse for companies choosing models for real work. Vals AI's benchmarks show a jagged frontier:... - [Why AI Won't Replace Stock Pickers](https://aietalks.com/talks/why-ai-wont-replace-stock-pickers): Jack Woodruff, Candlestick Capital & Vinoo Ganesh, Kepler. Jack Woodruff argues that AI will change the work around fundamental investing without removing the need for stock pickers. The central limit is human attention:... - [Your Agent Just Paid for Nothing: 6 Attacks on x402 Payments](https://aietalks.com/talks/your-agent-just-paid-for-nothing-6-attacks-on-x402-payments): Ishween Kaur, SoFi. Ishween Kaur explains x402 as a way for software to buy data and API responses with small on-chain payments. She maps the protocol across a... - [Your AI Agent Starts Every Session Like a New Hire](https://aietalks.com/talks/your-ai-agent-starts-every-session-like-a-new-hire): Killian Carlsen-Phelan, Sonar. Killian Carlsen-Phelan argues that coding agents lose time and tokens because each session starts without knowledge of the project's conventions, architecture, libraries, and existing code.... - [Catch the Bugs Your Code Review Misses](https://aietalks.com/talks/catch-the-bugs-your-code-review-misses): Killian Carlsen-Phelan & Taylor Luttrell-Williams, Sonar. Taylor Luttrell-Williams and Killian Carlsen-Phelan demonstrate Gitar, Sonar's code change verification agent, on a small Python ticketing application. The first pull request adds ticket transfers,... - [How Revolut Cut AI Coding Costs 90% With Self-Hosted Open Models](https://aietalks.com/talks/how-revolut-cut-ai-coding-costs-90-with-self-hosted-open-models): David Abitbol, Revolut. David Abitbol describes how Revolut evaluated AI coding tools across 2,000 engineers and moved much of its workload to self-hosted open models. Gemini initially reduced... - [We Trust AI Agents That Break Every Day](https://aietalks.com/talks/we-trust-ai-agents-that-break-every-day): Shy Ruparel, Temporal. Shy Ruparel presents four gaps between experimenting with AI agents and operating them in production. Engineers use agents daily at much higher rates than they... - [Why a One-Line Change Can Break Your Whole Stack](https://aietalks.com/talks/why-a-one-line-change-can-break-your-whole-stack): Cole Gannaway, Qodo. Cole Gannaway argues that software teams now need to improve the review and verification layer around increasingly fast code generation. He uses a one-line Bitwarden... - [Why Humans Can't Keep Up With AI-Written Code](https://aietalks.com/talks/why-humans-cant-keep-up-with-ai-written-code): Tariq Shaukat, Sonar. Tariq Shaukat compares agentic software development with the factory changes associated with Toyota and Lexus. Earlier software practices borrowed ideas such as quality at the... - [AI That Hunts Logic Bugs Before Attackers Do](https://aietalks.com/talks/ai-that-hunts-logic-bugs-before-attackers-do): Taylor Luttrell-Williams, Sonar. Taylor Luttrell-Williams explains why pattern-based SAST does not catch every security problem. Code can be valid and still violate the application's intended rules, such as... - [Exploits Now Beat Patches: Open Source Security in the AI Era](https://aietalks.com/talks/exploits-now-beat-patches-open-source-security-in-the-ai-era): Patrick Smyth, Chainguard. Patrick Smyth argues that frontier AI has changed the timing and scale of open source attacks. He cites Mandiant data showing that exploits now arrive... - [How Anthropic Merges 30,000 Code Changes a Day](https://aietalks.com/talks/how-anthropic-merges-30-000-code-changes-a-day): Manish Kapur, Sonar & Addy Osmani, Anthropic.
| – |
| 7 Nov 2026 |