Build hands-on expertise with PyTorch and open-weight LLMs as an Arkion Software Intern working remotely or hybrid in the Greater Toronto Area.

This Fall/Winter internship focuses on adapting large language models for security and identity work. You will prepare datasets, run fine-tuning jobs, and build evaluations with the founding team. The work reports to the Chief Product Officer and uses real models and GPUs.

Key Responsibilities:

- Collect, clean, and prepare training and evaluation datasets
- Fine-tune open-weight LLMs using SFT and LoRA
- Run and monitor GPU training jobs efficiently
- Build evaluation sets and task-focused benchmarks
- Compare runs and report measured results clearly

Requirements:

- PyTorch and Hugging Face Transformers, Datasets, and PEFT
- Hands-on experience with open-weight large language models
- Fine-tuning experience using SFT, LoRA, or QLoRA
- Dataset preparation and model evaluation experience
- Understanding of GPU memory, batch sizes, and mixed precision

Apply your LLM fine-tuning, dataset, evaluation, and GPU skills to Arkion’s security and identity work.

#J-18808-Ljbffr

📌 Software Intern LLM Fine-Tuning (Toronto)
🏢 Arkion Identity Systems
📍 Toronto