Stateful environments for vertical agents — Josh Purtell, Synth Labs

JPJosh PurtellIntervenant · Synth
Chaîne
AI Engineer
Publié
22 juil. 2025
Vidéo
6:50
Original
anglais
Ouvrir sur YouTube
L'article n'est pas encore prêtNous avons les informations de cette vidéo, mais son article n'a pas encore été rédigé. En attendant, vous pouvez la regarder sur YouTube.Ouvrir sur YouTube

Sur la même chaîne

7 diapositives18:56EN
AI Engineer· 27 oct. 2026John Feminella: Five Controls for AI Agents, and the $900 Million Bank ErrorJohn Feminella · Two SigmaA payment mistake shows why good intentions and approval prompts are not enough. Feminella describes practical controls for giving agents useful autonomy.Sécurité et gouvernanceIngénierie d’agentsIA en finance+2
5 diapositives18:12EN
AI Engineer· 27 oct. 2026Tim Tyler: An Accountability Framework for AI Agents as a Digital WorkforceTim Tyler · GraviteeThe Gravitee speaker applies identity, sponsorship, authority and oversight to agents, drawing on how organizations manage human work.Sécurité et gouvernanceIngénierie d’agentsAuthentification des agents+1
6 diapositives23:22EN
AI Engineer· 27 oct. 2026Noyan Tokgozoglu: How Optiver Builds an AI Platform for Fast, Measurable DecisionsNoyan Tokgozoglu · OptiverThe trading firm’s AI engineering lead explains same-day model adoption, on-premises infrastructure, cost controls and the limits of replaying agent evaluations.AI platform engineeringÉvaluationsAI cost & performance+2
6 diapositives16:38EN
AI Engineer· 27 oct. 2026Aiden Sochowski: Securing 300 Agents in a 60-Person CompanyAiden Sochowski · RunlayerRunlayer’s founding engineer explains what session monitoring revealed about agent permissions, prompt injection, tool errors and runaway cost.Ingénierie d’agentsAI securityPrompt injection+2
10 diapositives56:01EN
AI Engineer· 26 oct. 2026Xander Song on Evaluating Agents with Harbor and PhoenixXander Song · Arize AIArize AI engineer Xander Song explains how Harbor runs realistic agent tasks in isolated sandboxes, and how his team uses Phoenix to compare prompt changes, inspect trajectories, and test a SQL tool.Évaluations et observabilitéIngénierie d’agentsSandboxes & isolation+2
7 diapositives19:43EN
AI Engineer· 26 oct. 2026Jack Wotherspoon on Building a Team of Managed AgentsJack Wotherspoon · Google DeepMindGoogle DeepMind’s Jack Wotherspoon introduces Gemini Managed Agents: cloud sandboxes that let teams run coding and research agents with a single API call. He explains how persistent environments, reusable skills, proxy-injected credentials, schedules, and specialized agents fit together.Ingénierie d’agentsAgent orchestrationSandboxes & isolation+2