Senior AI Software Engineer

About the role

Leyton CognitX is the AI and technology advisory arm of Leyton, with a team of over 200 tech experts and data scientists. Its tech team builds the software used by its consultants and clients, including business platforms, a client portal and AI products, and is moving quickly into AI engineering with LangChain, LangGraph, agents and MCP. This role is for an experienced engineer who will act as the technical reference of the team, designing and building software end to end and bringing AI in through agents, RAG and MCP integrations.

Key responsibilities

• Design and build scalable software end to end, covering APIs, business logic, front end, databases and security.
• Build and run AI features in production: RAG pipelines, LLM agents, MCP servers, prompt and context engineering, evaluation and cost control.
• Deploy solutions on the cloud, including AI services and GPU-backed inference.
• Own production incidents: diagnose fast, coordinate the fix and run blameless postmortems.
• Set coding standards (SOLID, Clean Architecture, design patterns) and lead thorough code reviews.
• Define the testing strategy (unit, integration, e2e, AI evals) and keep CI/CD pipelines reliable.
• Mentor junior developers and interns through pairing, reviews and feedback.
• Follow the AI and software ecosystem, test new tools and share results with the team.
• Work with product, DevOps, security and business teams and explain technical choices in plain language.

Required profile

• Engineering degree or Master's in Computer Science or a related field, with 5+ years of software development experience.
• Experience shipping AI-powered features in production such as agents, RAG systems or MCP integrations.
• Experience leading or mentoring developers, staying calm under pressure during incidents, and choosing simple solutions that work.
• Clear communication in French and English.

Required skills

• Back-end with Python (FastAPI or Django) or PHP (Laravel) or TypeScript (NestJS), plus a modern front-end framework (React, Vue.js or Nuxt.js).
• SOLID principles, design patterns, dependency injection and refactoring.
• REST APIs, OAuth 2.0, SSO, JWT, WebSockets and OWASP security practices.
• PostgreSQL and Redis; MongoDB is a plus.
• Scalable architectures, caching and message queues (RabbitMQ, Kafka).
• Docker, Git, CI/CD (GitHub Actions, GitLab CI or Azure DevOps); Kubernetes and Terraform are a plus.
• OpenAI, Anthropic and open-source model APIs, LangChain/LangGraph, MCP, RAG, and vector databases (pgvector, Qdrant, Pinecone).
• Azure, AWS or GCP, and GPU infrastructure (vLLM, Ollama) are a plus.

What we offer

• A key role in a team shaping how Leyton uses AI, with ownership of technical and architecture decisions.
• Products used daily by Leyton's teams, with visible impact.
• An environment with weekly tech reviews and internal learning paths.
• A competitive package based on experience, in a culture of excellence, continuous improvement and work-life balance.