Teams at Concurrences

Evaluate AI outputs for antitrust accuracy across jurisdictions; build rubrics, annotation guidelines, and tests; design AI instructions; run structured legal experiments; own domains like EU merger control and cartels; write evaluation scripts using AI coding tools.

Build and maintain RAG infrastructure for hundreds of thousands of legal documents: chunking pipelines, vector DB integration, embedding strategy, semantic search, retrieval/reranking/LLM orchestration, evaluation tooling, API components, and deployable backend services. Ship multiple features quickly and collaborate with CTO and founder on architecture and domain-driven trade-offs.

Build and own end-to-end RAG and inference infrastructure: chunking, embeddings, retrieval optimization, vector DB schema/indexing, LLM integrations, agentic workflows, APIs (Python/FastAPI), and CI/CD/observability for a production legal research engine.