Software Engineer, Agent Harness & Evaluation: Python, TypeScript, LLM Agents & Evals | San Jose, CA (On-site) | Up to $350K + Equity

Sourced from bigwavedigital.com

Build and improve production AI agents that solve hardware engineering problems like chip design, debug, and verification. You will find agent failure modes, design rigorous evaluations, and build new agent capabilities across tool interfaces, memory, and multi-step orchestration. The role blends applied research, customer-facing problem solving, and production software engineering in Python and TypeScript.

• Identify agent failure modes and build evaluations measuring performance, reliability, and efficiency
• Build new agent capabilities including tool interfaces, context management, memory, and multi-agent orchestration
• Analyze traces and experiment results to distinguish real improvements from benchmark noise
• Work directly with hardware engineers and domain experts to translate problems into agent fixes
• Turn research experiments into maintainable, tested, production-grade software

• 8-15 years as a software engineer with serious experience on large-scale agent systems (staff-plus bar)
• Deep depth in LLM agents, orchestration, tool interfaces, memory, tracing, ML inference, and transformer architectures
• Production software engineering in both Python and TypeScript
• Strong customer-facing ability to understand needs and deliver technical solutions independently
• CS, EE, or similar degree from a top university
• US citizens and Green Card holders only; no sponsorship

• Experience or interest in EDA, semiconductor domains, RTL, or hardware-software systems
• Prior work at a major EDA vendor or on-device ML for consumer silicon