Mid-Training Research Engineer: Data & Knowledge Pipelines

Thinking Machines Lab Inc.; San Francisco (CA); Hybrid; USD 350,000 - 475,000; Full time

Benefits offered by this job

Health, dental, and vision benefits

Unlimited PTO

Parental leave

Relocation support

Job summary

Thinking Machines Lab Inc. is seeking an experienced Mid-Training Research Engineer to own data curation, synthetic generation, and quality pipelines for next-generation base models in SF. You will shape how models reason, recall, and perform across diverse domains, bridging research and practical engineering.

The role blends data-centric training, evaluation design, and scalable data pipelines, reporting to the Head of Research in a San Francisco-based team. Visa sponsorship is available.

Qualifications

• Proficiency in Python and familiarity with deep learning frameworks.
• Bachelor's degree or equivalent in CS, ML, Physics or Mathematics.
• Clarity in communication, ability to explain complex technical concepts in writing.

Responsibilities

• Own data: source, curate, synthesize data; build training data pipelines.
• Improve model knowledge with targeted corpora, synthetic rephrasings and QA over source documents.
• Instill behaviors and align mid-training with post-training and RL considerations.
• Build the quality pipeline: train classifiers and verifiers, calibrate data quality.
• Develop and tune the recipe: datasets, training stages, annealing schedules, hyperparameters.
• Iterate on evals to ensure evaluations capture meaningful improvements.
• Debug and understand results; scale and explore new data approaches.

Skills

Python

Deep learning frameworks (PyTorch / TF

Written communication

Education

Bachelor's degree or equivalent in CS/ML/Physics/Math

Tools

PyTorch

TensorFlow

JAX

Job description

Thinking Machines Lab Inc. is seeking an experienced Mid-Training Research Engineer to own data curation, synthetic generation, and quality pipelines for next-generation base models in SF. You will shape how models reason, recall, and perform across diverse domains, bridging research and practical engineering.

The role blends data-centric training, evaluation design, and scalable data pipelines, reporting to the Head of Research in a San Francisco-based team. Visa sponsorship is available.