Agent orchestration is so two-years ago

Agent orchestration is so two-years ago

Ryan welcomes Saahil Jain, CTO of You.com, to discuss why building agents with a 2024 mindset is a mistake as modern models improve at long-horizon tasks, why heavy orchestration layers can hurt model performance more than help it, and why the 2026 competitive edge actually comes from information retrieval and unique data paired with end-to-end evaluation.  Episode notes:  You.com [http://you.com] is an AI-powered search and productivity engine helping enterprises find information, create content, and automate complex tasks using web search APIs, multi-model AI access, and agentic intelligence.  Connect with Saahil on LinkedIn [https://www.linkedin.com/in/saahiljain/] or reach out to him at  saahil@you.com [saahil@you.com].  Today’s shoutout goes to user knittl [https://stackoverflow.com/users/112968/knittl] for winning a Populist badge on their answer to Remove all null values [https://stackoverflow.com/questions/39500608/remove-all-null-values]. See Privacy Policy at https://art19.com/privacy [https://art19.com/privacy] and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info [https://art19.com/privacy#do-not-sell-my-info].

The good, the bad, and the AI apps

Ryan welcomes Benny Chen, co-founder of Fireworks AI, to the show to explore what actually makes an AI application good or not, how to balance qualitative signals with quantitative metrics when evaluating AI, and how open-source eval protocols and community efforts are setting the standard for AI evaluation.  Episode notes:  Fireworks AI [https://fireworks.ai/] is a cloud platform designed for developers and enterprises to run, customize, and scale open-source generative AI models.  Connect with Benny on LinkedIn [https://www.linkedin.com/in/benny-yufei-chen-2238575a/]. Congrats to user techtabu [https://stackoverflow.com/users/1257729/techtabu] for winning a Stellar Answer badge for answering How can I delete all local Docker images? [https://stackoverflow.com/questions/44785585/how-can-i-delete-all-local-docker-images]. See Privacy Policy at https://art19.com/privacy [https://art19.com/privacy] and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info [https://art19.com/privacy#do-not-sell-my-info].

How do you turn AI coding chaos into a repeatable playbook?

Vivek Raghunathan, SVP of engineering at Snowflake, joins Leaders of Code at Snowflake Summit to break down the five-stage framework his org used to go from "let chaos reign" to a repeatable, org-wide system for AI-assisted engineering. Vivek explains how Snowflake systematically rolled out coding agents across its engineering org — starting with unrestricted experimentation, then codifying what worked into a shared vocabulary of 14 "AI design patterns," from plan-in-English to fencing off parallel agents to reducing on-call toil through continuously updated skills. Vivek walks through the "inner loop" and "outer loop" of software development, explains Snowflake's internal Yegge scale for measuring how far engineers have progressed along that continuum, and shares how a three-person team used coding agents to deliver a 40x improvement on Snowflake's query compiler. The discussion also: * Breaks down Snowflake's "focus weeks," where engineers get dedicated time to either catch up on best practices or push the frontier further. * Explores the pioneers/settlers/skeptics framework for meeting engineers where they are in adopting AI tools, and why the shift can trigger something like the stages of grief. * Covers how Snowflake cut release validation time from 15 days to a single day, and why more automated testing hasn't come at the cost of production stability. * Looks ahead to a four-step maturity model for on-call and incident response, where agents may eventually take primary on-call duty. Connect with Vivek Raghunathan on LinkedIn [https://www.linkedin.com/in/raghunathanvivek]. See Privacy Policy at https://art19.com/privacy [https://art19.com/privacy] and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info [https://art19.com/privacy#do-not-sell-my-info].

Why intent prediction needs more than an LLM

Ryan sits down with Frank Portman, CTO at Yobi, to talk about why next-token prediction, though great for language, isn’t the right inductive bias for forecasting human behavior. They discuss how Yobi builds a “foundation model of behavior” using transformers and graph neural networks instead of chat-style LLMs, and what it takes to run millions of personalization decisions per second while keeping consumer data private. Episode notes: Yobi [https://www.yobi.ai/] is a behavioral AI company building foundation models that predict future behavior for ad tech, marketing, and more. Connect with Frank via fportman.com [http://fportman.com] or at yobi.ai [http://yobi.ai]. Congrats to Hooked [https://stackoverflow.com/users/249341/hooked] on winning a Populist badge for their answer to Removing whitespace around a saved image [https://stackoverflow.com/questions/11837979/removing-white-space-around-a-saved-image]. See Privacy Policy at https://art19.com/privacy [https://art19.com/privacy] and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info [https://art19.com/privacy#do-not-sell-my-info].

Code isn’t the only thing causing your production failures

Ryan sits down with Anish Agarwal, CEO and co-founder of Traversal, to chat about why AI coding agents have made writing code easier but running it safely in production harder, why production failures are really caused by interactions between systems and not just the code itself, and how teams can troubleshoot more effectively when traditional observability tools are not enough for agentic AI workflows. Episode notes:  Traversal [https://www.traversal.com/] is an AI-powered autonomous SRE for complex software systems with automatic triage alerts, root cause investigation, and incident prevention at petabyte scale.  Connect with Anish on LinkedIn [https://www.linkedin.com/in/anish-agarwal-io/] or reach out to him at  anish@traversal.com [anish@traversal.com].  Our sixteenth Annual Developer Survey is now open and we want to hear your thoughts on all things software. Take the survey now [https://take.survey.stackoverflow.co/jfe/form/SV_4GHunpL3IfJ3rRc?utm_medium=podcast&utm_source=outreach&utm_campaign=dev-survey-2026&utm_content=launch-survey]! Congrats to user aioobe [https://stackoverflow.com/users/276052/aioobe] on winning a Populist badge for their answer to Javascript a=b=c statements [https://stackoverflow.com/questions/7511279/javascript-a-b-c-statements]. See Privacy Policy at https://art19.com/privacy [https://art19.com/privacy] and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info [https://art19.com/privacy#do-not-sell-my-info].