Distinguished Agent Engineer

Description

The company builds small, efficient language models designed to outperform large general-purpose models on the tasks that matter most. The team includes former members of Google Research, Apple, Stanford, and Cambridge.

The company's open source GLiNER models have been downloaded over 50M times, and teams at NVIDIA, Meta, and Airbnb use them.

The company has raised $25M, covered by TechCrunch, from Microsoft, Khosla Ventures, Insight Partners, GitHub CEO Thomas Dohmke, Docker CEO Scott Johnston, and others.

Responsibilities

• This is one of the most senior individual contributor engineering roles. You will set the technical direction for how agents are built on the company's models, write much of the core code yourself, and raise the standard for the people around you.
• Design and own the architecture of agent systems built on the company's small, fast models, covering planning, tool use, memory, and multi-agent orchestration.
• Determine where specialized models outperform frontier LLMs inside agent loops, and prove it with benchmarks and real customer workloads.
• Build the evals, telemetry, and feedback loops that show whether agents are improving or regressing, and why.
• Work with research to turn new model architectures into agent capabilities and help decide what research should focus on next.
• Make agents fast and reliable at scale, including low-latency inference, high-concurrency serving, and graceful handling of failures.
• Lead data and fine-tuning strategy for agent-specific tasks such as tool calling, extraction, and routing.

Requirements

• You have 5+ years of software engineering experience, including 3+ years building with LLMs or machine learning in production.
• You have shipped agent systems that real users depend on and have learned from the failures.
• You have deep experience across the stack, including prompt and context engineering, tool and API design, vector databases, Kubernetes, and inference serving.
• You understand model behavior well, including fine-tuning, evals, failure modes, and latency and cost tradeoffs.
• You have set technical direction across several teams, not only within a single project.
• You write clearly and can make complex systems easy to understand.
• Nice to have: experience building high-concurrency microservices for agent workloads.
• Nice to have: familiarity with GLiNER or other information extraction architectures.
• Nice to have: open source contributions, papers, or talks on agents or LLM systems.