CAREERS @ GENERIA

LLM Training & Fine-Tuning Research Engineer

Shawinigan, Québec

Hybrid

Your mission: lead the research, design, and experimentation aspects of LLM training, and drive SOTA research in pre-training techniques, supervised instructive fine-tuning and advanced reinforcement learning strategies.

GenerIA is a deep tech company with strong ambitions both technical and ethical. We already define eco-responsibility standards with the world's first carbon-neutral AIs, which are at the same time totally secured by design. Our Agenda for this year includes developing a new series of yet more frugal generative models with SOTA performance, and to debut a photonic quantum model together with industry leaders in the domain.

This certainly sounds serious but our commitment comes with principles. GenerIA is first and foremost a human adventure, whose success is ultimately measured by the personal and professional well-being of each and every one of us. If you can help us achieve our mission, if you recognize yourself in our Values and if you like both challenges and fun, let's talk.

Key Responsibilities

  • Pre-training & continuous learning: design and implement pre-training pipelines using techniques like Masked Language Modeling (MLM) and other unsupervised learning approaches
  • Fine-tuning & supervised instruction: develop fine-tuning protocols that integrate supervised signals for model refinement
  • Reinforcement Learning integration: experiment with and apply reinforcement learning techniques (PPO, GRPO, DPO, KTO, etc.) to further align models' outputs with desired behaviors
  • Algorithm development & experimentation: formulate and test novel training strategies that push the boundaries of LLM performance
  • Cross-disciplinary collaboration: work closely with data scientists and deployment engineers to ensure that training innovations translate effectively into production environments.

Requirements

  • Advanced degree (MS/Ph.D.) in Computer Science, Machine Learning, Artificial Intelligence, or a closely related field
  • Deep understanding of linear algebra, calculus, probability, and optimization methods
  • High proficiency in Python and familiarity with C and Rust
  • Ability to articulate complex ideas clearly, both in written publications and verbal presentations.

Benefits

  • Competitive salary and equity package
  • Flexible work arrangements
  • Health and dental insurance
  • Professional development budget
  • Regular team events and activities.

Apply for this opening

Your privacy is actively protected.

Our in-house application system keeps your application from being exposed.

YOU:
YOUR LINKS:
ADDITIONAL INFORMATION: