Mercor

About the job

Mercor connects elite creative and technical talent with leading AI research labs. Headquartered in San Francisco, our investors include Benchmark , General Catalyst , Peter Thiel , Adam D'Angelo , Larry Summers , and Jack Dorsey .

• Position: Human Baseliner for Open-Ended ML Research Tasks
• Type: Contract
• Compensation: $75–$90/hour
• Location: Remote
• Commitment: 20+ hours/week

Role Responsibilities

• Attempt open-ended machine learning research tasks under a fixed time and compute budget.
• Work independently in a sandboxed Linux environment with internet access.
• Use preferred tools, including IDEs and AI coding assistants like Cursor , Claude Code , and ChatGPT .
• Record full working sessions via screen recording.
• Complete pre-task and post-task questionnaires.
• Submit final work product, screen recording, and completed questionnaires for evaluation.

Qualifications Must-Have

• 3+ years of machine learning experience . Time in a PhD program counts.
• Attended a top-100 university or worked at FAANG or a comparable company .
• Experience with PyTorch , JAX , or TensorFlow .
• Deep expertise in at least one focus area: pretraining, PPO , reward shaping, fine-tuning, LoRA , RLHF , architecture design, contrastive training, generative modeling, multilingual experience, or data pipelines.

Required Domain Expertise

• Practical experience in Pretraining , Reinforcement learning , Post-training , Dataset curation , or Model architecture .

Logistics

• One baseline attempt per contractor per task.
• Each task may only be attempted once.
• All work is confidential and covered by NDA.
• Compute and environment are provided; no personal GPU required.