Fine-Tuning & Post-Training Engineer

About the Role

GoMeasure Platform runs goTalentOS — a specialised marketplace for AI workers, where engineers and domain experts don't only review and improve AI systems but also build and deploy them. We're looking for a Fine-Tuning & Post-Training Engineer to join this network on a flexible, contract basis. Adapt foundation models to a domain through fine-tuning, preference data and distillation. You'll work remotely, pick up engagements matched to your expertise, and be paid hourly for work that ships.

Key Responsibilities

• Run supervised fine-tuning and preference-optimisation cycles end to end.
• Design and curate training and preference datasets, including rater guidelines.
• Build reproducible training pipelines with tracked experiments.
• Distil large models into cheaper ones while holding quality on target tasks.
• Decide honestly when fine-tuning is the wrong answer versus prompting or retrieval.

Required Qualifications

Domain Expertise

• Hands-on fine-tuning or post-training of transformer models with measured outcomes.
• Strong data judgement — you know dataset quality dominates method choice.
• Able to report results with correct baselines and no cherry-picking.

Ways of Working

• Comfortable working independently and asynchronously in a remote, engagement-based setup.
• Strong written communication — decisions and trade-offs are recorded, not just made.
• Able to hand work over cleanly: another engineer can pick it up from your documentation.

How It Works

• Fully remote, contract engagements sized to your availability.
• Paid hourly at a rate set by role, seniority and verified skill level.
• Your skills are verified once through GoMeasure and reused across engagements.