Research, Post-Training Data

The Research, Post-Training Data role at Thinking Machines Lab is pivotal in refining AI systems to be more useful, safe, and collaborative for human users. This position bridges raw model intelligence with practical applications, focusing on enhancing AI through post-training processes. Thinking Machines Lab is dedicated to advancing collaborative general intelligence, aiming to provide accessible AI tools tailored to individual needs.

Key responsibilities include designing and executing data collection and synthesis strategies that combine human feedback, preference data, and synthetic examples to guide model behavior. The role involves developing scalable frameworks for high-quality human labeling and synthetic data generation, researching human preferences to improve AI reasoning and helpfulness, and iterating on evaluation metrics to measure the impact of post-training interventions.

Candidates should possess strong engineering skills with the ability to contribute to and debug complex codebases. Experience in data curation, human feedback, or synthetic data generation for large language models is essential. Proficiency in Python and familiarity with deep learning frameworks such as PyTorch, TensorFlow, or JAX are required. A bachelor's degree or equivalent experience in Computer Science, Machine Learning, Physics, Mathematics, or a related field is necessary, along with clear communication skills to explain complex technical concepts.

Thinking Machines Lab offers a collaborative environment where interdisciplinary work is valued, blending research, data operations, and technical implementation to advance human-centered AI systems. The company encourages continuous learning and provides opportunities to publish and present research that contributes to the broader AI community.