📢 New: get today's jobs on our WhatsApp Channel
Jobiglo

No results.

Software Engineer - Machine-Verified LLM Inference (Keystone Project)

EPFL · Lausanne

Mid 🇬🇧 English
Python C/C++ Rust OCaml CUDA Triton PTX vLLM SGLang PyTorch Lean Isabelle HOL

Job description

About the role

The Keystone project, an ARIA-funded collaboration between EPFL and Imperial College London, builds a formally-verified machine learning inference engine. It seeks a software engineer to turn verified research prototypes into a production-grade, high-performance inference engine. Prior experience with formal verification is welcome but not required.

Key responsibilities

  • Bring systems programming and performance engineering expertise to support the project's research, collaborating with the EPFL and Imperial College London teams
  • Design, implement and maintain core components of the verified LLM inference engine, including the runtime and glue code linking verified code, GPU kernels and drivers
  • Manage engineering infrastructure: differential testing against reference engines (vLLM, SGLang), continuous integration for code and proofs, and performance benchmarking
  • Develop and maintain agentic AI pipelines for specification autoformalization, proof generation and proof repair
  • Write documentation and procedures to ensure reproducibility, and diagnose, prevent and repair failures and regressions
  • Analyse the security level and trusted computing base of the components, and contribute to red/blue team exercises within the ARIA programme
  • Contribute to open-source releases and engage with their user communities

Required profile

  • Degree in computer science or an equivalent field
  • Experience in the field
  • Mastery of English (oral and written); French is an asset but not required
  • Sense of priorities, integrity and autonomy, and good communication with proof engineers and systems researchers

Required skills

  • Systems programming and strong coding in Python, C/C++, Rust, OCaml or another functional language
  • One of: GPU programming (CUDA, Triton, PTX) or high-performance computing; ML inference or serving systems (vLLM, SGLang, PyTorch internals); interactive theorem proving (Rocq, Lean, HOL, Isabelle) or other formal methods
  • Experience with build systems, CI and test infrastructure; LLM-based development tools or agentic workflows are a plus

What we offer

  • A place in a dynamic team on a high-profile ARIA-funded project, in a multicultural academic environment
  • Continuing education and professional development opportunities
  • Funded travel for collaboration between Lausanne and London and for conferences

Applications are submitted only through the online platform, with a cover letter, CV and three referees. The contract starts on 1 December 2026, is a one-year fixed-term position renewable according to project duration, and runs at full activity rate.

Questions fréquentes

Le salaire n'est pas communiqué publiquement par le recruteur. Vous pouvez postuler et négocier directement avec EPFL.
Cliquez sur "Postuler maintenant" en haut de la page. Vous pouvez importer votre CV en 1 clic — Jobiglo extrait automatiquement vos informations et postule pour vous.
Source : ats:successfactors

Why are you reporting this job?

Thank you for your report. We will review this job.

Apply in 30 seconds

Enter your email to apply. An account will be created automatically.

Apply now →

By continuing, you accept our terms of use.

Already have an account? Login

A question about this job?

Ask it here: you will get the full job summary by e-mail, right away.

Published 2 weeks ago

Expires 1 month from now

25 views · 0 interested

Boost your chances

Upload your CV — we will match you with relevant openings.

Analyzing your CV...

EPFL

Lausanne