← back

Scaling AI Agents Without Breaking Reliability — Preeti Somal, Temporal

3.3K views · Jul 28, 2025 · 15:00 min · Watch on YouTube ↗
Takeaway

Temporal's decade-old durable workflow engine maps cleanly onto agent reliability needs and is now seeing Python SDK overtake others as agents go to production.

Summary

  • Preeti Somal (Temporal engineering) pitches Temporal as the durable-execution substrate for agents — mascot is a tardigrade for max reliability.
  • Agents are complex distributed systems: must orchestrate workflows, handle long-lived state, human approvals, parallel execution, and unreliable LLM/tool calls; Temporal has solved these for 10+ years in mission-critical apps.
  • Python overtook other SDK languages in January, reflecting AI-agent adoption; customers include Dust, Gorgias (customer service for Reebok, Timbuktu, Glossier), and payments use-cases.
  • Core pitch: outsource reliability/scalability so engineers focus on business logic; agents inherit guardrails, retries, and observability from Temporal's battle-tested platform.
temporaldurable-executionreliability
Original description
As AI agents move from prototypes to production, developers are running into new challenges with orchestration, failure handling, and infrastructure. This session will unpack lessons from teams already building real-world systems and share how to design for reliability from the start.

About Preeti Somal
Preeti is Senior Vice President of Engineering at Temporal. Preeti is passionate about building great products, growing world class organizations and solving complex problems. Prior to Temporal, Preeti led the Platform, Security and IT engineering organizations at HashiCorp. Her extensive career includes engineering leadership roles at Yahoo!, VMware and Oracle. While at Yahoo! Preeti was VP of Cloud Services in the Platform organization delivering highly scalable services used by engineers across Yahoo to build and operate applications with improved agility, reliability and security. These services power Yahoo!’s consumer and advertising business.

Recorded at the AI Engineer World's Fair in San Francisco. Stay up to date on our upcoming events and content by joining our newsletter here: https://www.ai.engineer/newsletter