Verified Mercor opportunity
We're looking for experienced machine learning researchers with hands-on experience training and improving language models end-to-end. You'll work on well-scoped empirical open-ended LLM research problems. • * * — Responsibilities • Train transformer-based language models from scratch and fine-tune open-weight models. • Get the most out of limited data and compute. • Construct training corpora from raw web-scale sources. • Build post-training pipelines. • Diagnose and resolve training issues. • * * — Requirements — We are looking for candidates with strong expertise in one or more of the following areas: — Foundation Model Pre-training — Experience with: • Training transformer-based language models from scratch, end-to-end. • Data- and compute-constrained regimes: allocating a fixed budget across model size, tokens, and epochs. • Diagnosing optimisation failures, convergence issues, and training instabilities. — Pre-training Data — Experience with: • Corpus construction from raw web crawls and other large unfiltered sources. • Data filtering, deduplication, quality classification, and mixture/ordering optimisation. • Measuring data interventions rigorously. — LLM Post-Training — Hands-on experience with one or more of: • Supervised fine-tuning, including building your own datasets via synthetic generation, noisy or weak supervision, and rejection sampling. • Preference optimisation (DPO, RLHF, RLAIF) and reward modelling / human-preference prediction. • Alignment fine-tuning: shaping refusal behaviour, truthfulness, and unbiased reasoning while preserving general capability. • Fine-tuning for narrow, verifiable domains (math, code, games, structured prediction) where outputs can be checked programmatically. — Additional Areas of Interest — Experience in any of the following is a plus: • Scaling laws and training-efficiency research. • Curriculum learning and data ordering. • LLM evaluation: benchmark construction, contamination control, statistically sound comparisons. • Reinforcement learning for language models. • Model alignment and AI safety. — General Qualifications • 3+ years of machine learning research experience (PhD research counts toward this requirement). • Strong experience with PyTorch, JAX, TensorFlow, or similar ML frameworks. • Degree from a top-100 university, experience at a FAANG or comparable AI company, or an equivalent research track record through publications or impactful open-source contributions. • * * — Why Join • Work on cutting-edge foundation model research. • Collaborate with leading AI researchers on challenging, high-impact projects. • Flexible, project-based work with competitive compensation.