Software Engineer - Backend & GenAI

About the role

Lextract is looking for a Backend Software Engineer with a focus on Generative AI to design and scale the AI core of its legal due‑diligence platform. You will work at the intersection of large‑language‑model pipelines, agentic workflows, and robust backend infrastructure to transform how M&A lawyers process documents.

Key responsibilities

• Architect and scale AI systems that power the end‑to‑end legal due‑diligence pipeline.
• Build multi‑approach retrieval (RAG, graph‑based) and deep per‑document and cross‑document analysis across diverse formats.
• Design source‑referencing mechanisms ensuring full traceability of insights for lawyers.
• Develop secure, high‑performance APIs and data infrastructure handling large document volumes.
• Define and implement evaluation metrics for complex, domain‑specific AI outputs.

Required profile

• Minimum 3 years of full‑time backend engineering experience with strong Python skills.
• Hands‑on experience with large language models, retrieval‑augmented generation, prompting and agentic patterns.
• Proficiency with cloud platforms (Azure, AWS or GCP) and container orchestration (Docker, Kubernetes).
• Zero‑trust security mindset, up‑to‑date with authentication, secure cloud and data‑protection best practices.

Required skills

• Python
• FastAPI
• PostgreSQL
• Next.js, React, TypeScript
• LLM pipelines, RAG pipelines, agentic frameworks
• Docker, Kubernetes, Terraform
• Azure (and familiarity with AWS/GCP)
• Zero‑trust security principles

What we offer

• Fast‑moving early‑stage team with deep legal‑tech expertise.
• Competitive compensation and equity options.
• On‑site work from our Munich office, providing close collaboration with the team.
• End‑to‑end ownership of features with minimal dependencies.