SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Temporal is an open-source programming model that simplifies code, improves application reliability, and helps developers ship features faster. The company powers AI leaders including OpenAI, NVIDIA, Cursor, Lovable, and Replit, with expanding adoption across AI agents, data pipelines, and related use cases.
The AI Foundations team is tasked with accelerating Temporal adoption across the ecosystem by combining deep use-case understanding with rigorous computer systems and software design principles. In this Staff-level role, you will lead agent optimization efforts, designing tools and mechanisms to help Temporal users build agents optimized for token spend and response time while maintaining result quality.
You will own the technical direction for model routing and related optimization techniques including multi-agent architecture, cache policy, and context management. This requires direct, hands-on experience with AI agent development and optimization challenges. You will collaborate closely with other AI Foundations team members who focus on agentic development, ecosystem integrations, and policy/security systems.
Key responsibilities include: designing and building trustworthy agentic coding systems; developing deep expertise in AI application development patterns and emerging architectures; taking end-to-end ownership of new features with exceptional reliability and developer experience; working across Python, TypeScript, Java, and Go; serving as a domain expert on AI design patterns; collaborating with the developer community on debugging and feature feedback; writing technical documentation; and occasionally supporting customers with expert engineering guidance.
Temporal is fully remote and values action-oriented execution balanced with thorough technical grounding. Most work is open source. The role involves occasional travel to meet coworkers (1-2 weeks annually) and optional conference speaking.
