Context Engineer

About the Role

We're seeking a Context Engineer to join our partner companies solving the hardest problem in applied AI: getting the right information to models at the right time. This role — formally defined by Gartner in 2026 — goes beyond prompt engineering to build the systems that feed context to LLMs. You'll design retrieval pipelines, build memory architectures, optimize chunking strategies, and ensure AI agents have the grounded, relevant information they need to be accurate and useful. If prompt engineering was writing the question, context engineering is building the library.

What You'll Do

• Design and build RAG architectures with advanced retrieval strategies
• Implement document processing pipelines: parsing, chunking, embedding, indexing
• Build and optimize vector search systems for semantic retrieval
• Design memory architectures for conversational and agentic AI systems
• Implement context window management, summarization, and compression strategies
• Evaluate and benchmark retrieval quality with systematic testing

Requirements

• 4+ years of software engineering, 2+ working with LLMs and retrieval systems
• Deep experience with vector databases (Pinecone, Weaviate, pgvector, Qdrant, Chroma)
• Strong understanding of embedding models, chunking strategies, and reranking
• Experience with RAG frameworks (LlamaIndex, LangChain) and their retrieval patterns
• Knowledge of document parsing, OCR, and unstructured data processing
• Excellent English communication skills (B2+)

Nice to Have

• Experience with knowledge graph construction and hybrid search
• Background in information retrieval or search engineering
• Experience building context systems for production AI agents
• Knowledge of evaluation frameworks (RAGAS, DeepEval) for retrieval quality

Tech Stack

Benefits & Perks