AI Engineer Student

About the role#

Mawer is looking for an AI Engineer Student to join our team for the Winter 2027 term. You will work on the model layer of our internal AI platform, focusing on training, adapting, and evaluating models that we run on our own hardware. This is a full-time, paid position based in Calgary, Alberta, with a hybrid work structure.

What you'll do#

• Serve open-weight models locally and characterize them, including memory footprint, tokens per second, latency, and quality.
• Build and maintain training datasets from internal and public corpora while managing licensing and quality.
• Run fine-tuning experiments end to end and document results for reproducibility.
• Assist with reinforcement learning experiments, including reward design, preference data collection, and policy evaluation.
• Design evaluations that separate capability gains from benchmark overfitting.
• Contribute to team practices for model training and evaluation standards.

What you'll need#

• Currently enrolled in a post-secondary program in Computer Science or Computer Engineering and returning to full-time studies or graduating after the term.
• Hands-on experience running local models using tools like llama.cpp, vLLM, Ollama, or Hugging Face Transformers.
• Coursework or serious self-study in transformers and modern architectures, including attention, positional encoding, and mixture-of-experts.
• Proficiency in Python and PyTorch, with the ability to read model source code.
• Experience training or fine-tuning models, even at a small scale.
• Ability to maintain data quality and clear technical writing skills.
• Eligibility to work in Canada.
• Preferred experience includes reinforcement learning (DPO, RLHF), parameter-efficient fine-tuning (LoRA), distributed training, and inference optimization.

Location & details#

• Location: Calgary, Alberta, Canada
• Term: Winter 2027 (January to April)
• Modality: Hybrid