Member of Technical Staff | Agentic AI

The role

This person will design and build the architecture for a scalable agent platform. They will develop agent-driven workflows and systems for complex investment problems using frontier models and production engineering.

What you will do

• Design and build the architecture behind a scalable, reliable agent platform
• Design and build agent-driven workflows and systems for complex investment problems
• Develop evaluation, observability and feedback systems
• Solve infrastructure and deployment challenges of running agents reliably in production

Required qualifications

• Built agent systems at a frontier AI lab, agentic startup, top fund, big tech company, or similar
• Deployed agents in production
• Experience with orchestration, tool use, context management and grounding
• Understand agent failure modes and have built evaluations and observability
• Understand financial markets and have a strong interest in investing
• BS/MS/PhD in Computer Science or a related field, or equivalent practical experience

Nice to have

• Experience working with unstructured financial data and data providers (Bloomberg, Capital IQ, Refinitiv)
• Experience with agent safety, sandboxing or guardrails in high-stakes environments

Benefits

• Medical coverage
• Dental coverage
• Vision coverage
• Catered lunches
• Commuter benefits
• Relocation allowance available

Before you apply

• Current legal authorization to work in the US required
• Continuing work visa sponsorship available for full-time employees

About the company

Poesis is an AI-native investment firm running autonomous agents that predict markets, construct portfolios, and manage risk. The founders managed institutional capital at Capital Group and led enterprise ML at Goldman Sachs and Amazon.