AI Engineer (Brilliant Harvest)

Role overview Build and ship AI features for a platform used in an established industry, with a focus on agents that plan, make decisions, use tools, and complete multi-step tasks. The role spans agent development, retrieval-augmented generation (RAG), document processing, and model integrations, from design through production support. You’ll work with an AI team, take ownership of defined features, and grow toward more complex problems.

Responsibilities - Develop production agents, including orchestration, memory, tool integrations, safeguards, and human review for consequential actions. - Build and maintain RAG, language model, and document ingestion components, including parsing, chunking, enrichment, and indexing. - Diagnose answer and agent failures across tools, retrieval, and source documents. - Create evaluation data and automated checks to track task success, tool accuracy, cost, latency, and regressions. - Prototype models and frameworks, and contribute to design, code review, documentation, and delivery planning. - Coordinate with product, data, design, and QA partners to integrate AI features into the product.

Requirements - At least 3 years in software engineering or data science, including 1 year building AI, machine learning, or language model applications. - Practical experience with RAG, embeddings, vector search, or document ingestion, ideally in a production feature. - Strong Python skills and working experience in TypeScript, C#/.NET, or another modern backend language. - Working knowledge of API design, microservices, and cloud infrastructure. - Clear communication and a collaborative approach to planning, feedback, and raising blockers.

Nice to have - Experience with frontend frameworks such as React or React Native, prompt engineering, AI-assisted development tools, or agent and LLM evaluation methods. - Startup experience or familiarity with agriculture, heavy equipment, or dealership operations. - Experience building reliable agents for real users, including handling poor context, tool failures, and recovery.

Benefits and work setup - Remote-first, flexible work environment with mentorship and opportunities to take on broader work. - Competitive compensation, employee stock ownership, and benefits.