Forward Software Engineer
Mistral AI · Paris, Île-de-France
ABOUT MISTRAL
Mistral provides full-stack AI solutions: from frontier models to developer tools, applications, and compute. We partner with enterprises tackling the hardest problems across high-stakes industries like finance, manufacturing, defense, healthcare, and the public sector, co-creating customized AI systems that they can run on their terms.
We are a dynamic, collaborative team passionate about AI and its potential to transform society. Our diverse workforce thrives in competitive environments and is committed to driving innovation. Our teams are distributed between Europe, North America, Asia and the Middle East. We are creative, low-ego and team-spirited.
ABOUT THE JOB
As a Forward Software Engineer, you will build the systems behind some of the hardest problems Mistral takes on.
You join our team of engineers in Applied AI. We gravitate towards hard problems, decide how to solve them, and ship fast. You work closely with our customer-facing teams, and what works goes into production with our customers and across the company. When a project needs it, you work directly on customer engagements.
We're looking for a strong software engineer who has shipped agentic or GenAI systems to production. You've felt the pain of agents that work in the demo and fall apart on real tasks: tool calls that fail silently, loops that never end, and quality that drifts after launch. That production depth is your edge. You build across the stack, and you'd rather ship a plain UI on a solid foundation than a polished UI on a fragile one.
You think like a founder. You start from the customer's problem, not the technology. You measure success by the value you deliver, and you care about building experiences people love to use. You like working at the edge of what AI can do today, and you push past it.
Your scope is end to end, from first prototype to production. You help set the standard for how the team builds agents that are reliable, observable, and safe to run against sensitive systems.
Security is a big part of where we're heading. If you're drawn to it, that's a strong plus. You might come from security with a real interest in AI engineering, or from AI engineering with a real interest in security. Both paths fit. Curiosity about how systems break, and a habit of learning fast from the people who know the domain, matter. Experience at a security company counts too.
WHAT YOU WILL DO
- Build AI systems end to end, from first prototype to production.
- Design the agent architecture: orchestration, tool use, memory, retrieval, and the interfaces between agents and the systems they act on.
- Build across the stack, from backend services and APIs to the interfaces people use to work with agents.
- Turn what works for one customer into patterns and platforms that work across many, without rebuilding each time.
- Make systems hold up under real load: throughput, latency, cost, and failure recovery are part of the design from the start.
- Build security into agents from day one: permissions, sandboxing, prompt-injection defenses, and audit trails for agents that touch sensitive systems.
- Build the evaluation loop: task-level evals, tracing, and checks that catch agent failures and quality drift before customers do.
- Work with researchers, engineers, and customer-facing teams to turn hard problems into systems people rely on.
WHAT WE'RE LOOKING FOR
- You have 5+ years building production software, with at least one agentic or GenAI system shipped to production that you ran after launch.
- You have full ability to ship across the stack: proficient NodeJS e.g. NextJS/Express/Fastify or similar or Python e.g. FastAPI. You are confident with TypeScript + React/NextJS or similar. Solid beats polished.
- You have hands-on depth in agent design: orchestration, tool use, retrieval, evaluation, and the ways agents fail in the real world.
- You work autonomously: you find the problem, decide the approach, and own the outcome.
- You communicate clearly with technical and non-technical audiences, and are fluent in English.
- You ship in increments and are obsessed with time to value
- You have product judgment: you learn what users need through customer-facing teams, cut scope hard, and care how the result feels to use.
IDEALLY YOU HAVE
- Open-source contributions, particularly in agents or LLMs.
- Experience deploying and running what you build: Docker, Kubernetes, Terraform, and AWS, GCP, or Azure.
- A security background in application, cloud, or AI security, such as prompt injection, securing agents, or supply chain (SBOM, SLSA). If that's you, flag it in your application.
- Experience building for security workflows, such as alert triage, threat hunting, or investigation agents, or with SIEM, XDR, or SOAR tooling.
HOW WE WORK IN APPLIED AI
- We care about people and outputs.
- What matters is what you ship, not the time you spend on it
- Bureaucracy is where urgency goes to vanish. You talk to whoever you need to talk to. The best idea wins, whether it comes from a principal engineer or someone in their first week.
- Always ask why. The best solutions come from deep understanding, not from copying what worked before
- We say what we mean. Feedback is direct, timely, and given because we care.
- No politics. Low ego, high standards.
- We embrace an unstructured environment and find joy in it.
WHAT WE OFFER
We offer a comprehensive benefits package designed to support your well-being, growth, and work-life balance. Benefits vary by country and may include healthcare coverage, parental leave, retirement plans, relocation support, wellness programs, meal and transportation allowances, and other location-specific perks.
For the most up-to-date details on benefits available in your location, please refer to our Benefits page https://app.notion.com/p/mistralai/Benefits-at-Mistral-36e6ba59a7fe836b93dd01737fcc27ef?source=copy_link.
PRIVACY POLICY
Your privacy matters to us. You can learn more about how we handle your personal data in our Applicant Privacy Policy https://legal.mistral.ai/terms/applicant-privacy-policy.