AI Engineer (Core Harness)

The role

An AI Engineer on the Core Harness team will work across model behaviour and the systems that shape it. You will build the systems that turn models into useful, reliable agents including the agent loop, tools, memory, context management and execution environment.

What you will do

• Own the development and application of LLM-based features end-to-end
• Build and maintain evaluation frameworks for prompt iteration and semantic retrieval
• Optimize embeddings and vector stores for performance and scalability
• Interact with users to understand their problems and design solutions
• Stay up-to-date with the latest trends in LLMs and applied AI practices
• Collaborate with a cross-functional team of engineers, product managers, designers, and other members

Required qualifications

• Engineering experience at tech and product-driven companies
• Shipping multi-provider LLM-based solutions to production (using e.g. OpenAI, Anthropic, Google, etc.)
• Proficiency with LLM infra platforms (prompt management, logging/tracing, evals)
• Experience designing large-context LLM systems (RAG, knowledge graphs, hybrid search, memory)
• Building features end-to-end with TypeScript, React.js, and Node.js

Nice to have

• Experience working in a startup or being drawn to the zero-to-one phase

Benefits

• Salary slightly above market
• Equity above market
• Relocation assistance

About the company

Granola is building tools in an exciting time for tool builders. The team works in person five days per week in an office at Old Street in London.