About the Role

Thinking Machines builds multimodal-first. We’re looking for new team members to advance the science of visual perception and multimodal learning. We think about how vision and language interact at scale. We design architectures that fuse pixels and text, build datasets and evaluation methods that test real-world comprehension, and develop representations that let models ground abstract concepts in the physical world. Our goal is to create multimodal systems that support seamless integration into real-world environments.

You’ll work at the intersection of visual understanding, multimodal reasoning, and large-scale model training. You’ll help develop the architectures, data, and evaluation tools that teach AI to see, understand, and collaborate. The best candidate is curious about multimodal interfaces, has experience running large scale experiments and is comfortable contributing to complex engineering systems. While we are looking for a person with expertise in multimodality, Thinking Machines Lab operates in a unified fashion and expects new hires to work across modalities as one team.

This role blends fundamental research and practical engineering, as we do not distinguish between the two roles internally. You will be expected to write high-performance code and read technical reports. It’s an excellent fit for someone who enjoys both deep theoretical exploration and hands-on experimentation, and who wants to shape the foundations of how AI learns.

Note: This is an "evergreen role" that we keep open on an on-going basis to express interest in this research area. We receive many applications, and there may not always be an immediate role that aligns perfectly with your experience and skills. Still, we encourage you to apply. We continuously review applications and reach out to applicants as new opportunities open. You are welcome to reapply if you get more experience, but please avoid applying more than once every 6 months. You may also find that we put up postings for singular roles for separate, project or team specific needs. In those cases, you're welcome to apply directly in addition to an evergreen role.

What You’ll Do

• Own research projects on training and performance analysis of multimodal AI models.
• Curate and build large-scale datasets and evaluation benchmarks to advance vision capabilities.
• Work with our data infrastructure engineers, pretraining researchers and engineers, and product team to create frontier multimodal models and the products that leverage them.
• Publish and present research that moves the entire community forward. Share code, datasets, and insights that accelerate progress across industry and academia.

Skills and Qualifications

Minimum qualifications:

• Ability to design, run, and analyze experiments thoughtfully, with demonstrated research judgment and empirical rigor.
• Understanding of machine learning fundamentals, large-scale training, and distributed compute environments.
• Proficiency in Python and familiarity with at least one deep learning framework (e.g., PyTorch, TensorFlow, or JAX). Comfortable with debugging distributed training and writing code that scales.
• Bachelor’s degree or equivalent experience in Computer Science, Machine Learning, Physics, Mathematics, or a related discipline with strong theoretical and empirical grounding.
• Clarity in communication, an ability to explain complex technical concepts in writing.

Preferred qualifications — we encourage you to apply even if you don’t meet all preferred qualifications, but at least some:

• Research or engineering contributions in visual  reasoning, spatial understanding, or multimodal architecture design.
• Experience developing evaluation frameworks for multimodal tasks.
• Publications or open-source contributions in vision-language modeling, video understanding, or multimodal AI.
• A strong grasp of probability, statistics, and ML fundamentals. You can look at experimental data and distinguish between real effects, noise, and bugs.
• PhD in Computer Science, Machine Learning, Physics, Mathematics, or a related discipline with strong theoretical and empirical grounding; or, equivalent industry research experience.

Logistics

• Location: This role is based in San Francisco, California. 
• Compensation: Depending on background, skills and experience, the expected annual salary range for this position is $350,000 - $475,000 USD.
• Visa sponsorship: We sponsor visas. While we can't guarantee success for every candidate or role, if you're the right fit, we're committed to working through the visa process together.
• Benefits: Thinking Machines offers generous health, dental, and vision benefits, unlimited PTO, paid parental leave, and relocation support as needed.

About the Role

The role of pre-training researchers sits at the core of our roadmap. This work advances the science of how large models learn from data. You’ll explore new pre-training methods, architectures, and learning objectives that make model training efficient, robust, and aligned with human goals.

This role blends fundamental research and practical engineering, as we do not distinguish between the two roles internally. You will be expected to write high-performance code and read technical reports. It’s an excellent fit for someone who enjoys both deep theoretical exploration and hands-on experimentation, and who wants to shape the foundations of how AI learns.

Note: This is an "evergreen role" that we keep open on an on-going basis to express interest in this research area. We receive many applications, and there may not always be an immediate role that aligns perfectly with your experience and skills. Still, we encourage you to apply. We continuously review applications and reach out to applicants as new opportunities open. You are welcome to reapply if you get more experience, but please avoid applying more than once every 6 months. You may also find that we put up postings for singular roles for separate, project or team specific needs. In those cases, you're welcome to apply directly in addition to an evergreen role.

What You’ll Do

• Research and develop new methodologies for pre-training.
• Work in areas such as scaling, architecture, algorithms, or optimization of large scale training runs depending on your research interest and experience.
• Design data curricula and sampling strategies that improve learning dynamics and model generalization.
• Collaborate with infrastructure and data teams to conduct large-scale experiments efficiently and reproducibly.
• Publish and present research that moves the entire community forward. Share code, datasets, and insights that accelerate progress across industry and academia.

Skills and Qualifications

Minimum qualifications:

• Ability to design, run, and analyze experiments thoughtfully, with demonstrated research judgment and empirical rigor.
• Experience with distributed or high-performance computing environments.
• Proficiency in Python and familiarity with at least one deep learning framework (e.g., PyTorch, TensorFlow, or JAX). Comfortable with debugging distributed training and writing code that scales.
• Bachelor’s degree or equivalent experience in Computer Science, Machine Learning, Physics, Mathematics, or a related discipline with strong theoretical and empirical grounding.
• Clarity in communication, an ability to explain complex technical concepts in writing.

Preferred qualifications — we encourage you to apply even if you don’t meet all preferred qualifications, but at least some:

• A strong grasp of probability, statistics, and ML fundamentals. You can look at experimental data and distinguish between real effects, noise, and bugs.
• Prior experience training or analyzing large-scale models, or contributing to pre-training or foundation model research.
• Strong publication record or open-source contributions in representation learning, optimization, scaling laws, or other areas of pre-training.
• Familiarity with curriculum learning, data selection, or active learning techniques.
• Experience designing or maintaining evaluation frameworks for large models.
• Contributions to open datasets, research publications, or data tooling.
• PhD in Computer Science, Machine Learning, Physics, Mathematics, or a related discipline with strong theoretical and empirical grounding; or, equivalent industry research experience.

Logistics

• Location: This role is based in San Francisco, California. 
• Compensation: Depending on background, skills and experience, the expected annual salary range for this position is $350,000 - $475,000 USD.
• Visa sponsorship: We sponsor visas. While we can't guarantee success for every candidate or role, if you're the right fit, we're committed to working through the visa process together.
• Benefits: Thinking Machines offers generous health, dental, and vision benefits, unlimited PTO, paid parental leave, and relocation support as needed.

About the Role

The role of pre-training researchers sits at the core of our roadmap. This work blends research with large-scale data engineering to help assemble the pre-training datasets and data systems that underpin the next generation of AI models. You’ll design and implement methods for sourcing, curating, and analyzing pre-training data for quality and performance.

You’ll work with automated pipelines and human-in-the-loop processes, contributing both scientific insight and production-grade code. It’s ideal for someone who enjoys working at the intersection of data, machine learning, and systems, and who’s excited by the challenge of shaping frontier AI.

This role blends fundamental research and practical engineering, as we do not distinguish between the two roles internally. You will be expected to write high-performance code and read technical reports. It’s an excellent fit for someone who enjoys both deep theoretical exploration and hands-on experimentation, and who wants to shape the foundations of how AI learns.

Note: This is an "evergreen role" that we keep open on an on-going basis to express interest in this research area. We receive many applications, and there may not always be an immediate role that aligns perfectly with your experience and skills. Still, we encourage you to apply. We continuously review applications and reach out to applicants as new opportunities open. You are welcome to reapply if you get more experience, but please avoid applying more than once every 6 months. You may also find that we put up postings for singular roles for separate, project or team specific needs. In those cases, you're welcome to apply directly in addition to an evergreen role.

What You’ll Do

• Design and implement techniques for curating, sourcing, and filtering large-scale text, code, and multimodal data.
• Develop data quality metrics and analysis to measure coverage, diversity, and representativeness across sources.
• Collaborate with research and infrastructure teams to scale data processing systems efficiently and reproducibly.
• Investigate and mitigate data risks, including privacy, safety, and licensing concerns, to ensure responsible and ethical data use.
• Continuously evaluate dataset improvements by analyzing their downstream effects on model learning and behavior.
• Publish and present research that moves the entire community forward. Share code, datasets, and insights that accelerate progress across industry and academia.

Skills and Qualifications

Minimum qualifications:

• Proficiency in Python and familiarity with at least one deep learning framework (e.g., PyTorch, TensorFlow, or JAX). Comfortable with debugging distributed training and writing code that scales.
• Bachelor’s degree or equivalent experience in Computer Science, Machine Learning, Physics, Mathematics, or a related discipline with strong theoretical and empirical grounding.
• Clarity in communication, an ability to explain complex technical concepts in writing.

Preferred qualifications — we encourage you to apply even if you don’t meet all preferred qualifications, but at least some:

• A strong grasp of probability, statistics, and ML fundamentals. You can look at experimental data and distinguish between real effects, noise, and bugs.
• Experience with curation, preprocessing, and analysis of large-scale text, code, or multimodal datasets.
• Prior experience in data engineering, dataset construction, or large-scale web data processing for machine learning models.
• Experience evaluating or improving training data quality and knowledge of data ethics, safety, and licensing frameworks relevant to AI dataset creation.
• Contributions to open datasets, research publications, or data tooling.
• PhD in Computer Science, Machine Learning, Physics, Mathematics, or a related discipline with strong theoretical and empirical grounding; or, equivalent industry research experience.

Logistics

• Location: This role is based in San Francisco, California. 
• Compensation: Depending on background, skills and experience, the expected annual salary range for this position is $350,000 - $475,000 USD.
• Visa sponsorship: We sponsor visas. While we can't guarantee success for every candidate or role, if you're the right fit, we're committed to working through the visa process together.
• Benefits: Thinking Machines offers generous health, dental, and vision benefits, unlimited PTO, paid parental leave, and relocation support as needed.

About the Role

The role of post-training researchers sits at the core of our roadmap. This is the critical bridge between raw model intelligence and a system that is actually useful, safe, and collaborative for humans.

Post-training data research work sits at the intersection of human insight and machine learning. Our work combines human and synthetic data techniques, along with other innovative approaches, to capture the nuances of human behavior and use them to steer models. We research and model the mechanisms that create value for people to explain, predict, and optimize for human preferences, behaviors, and satisfaction. Our goal is to turn research ideas into data by scoping well-run data labeling or collection campaigns, and understanding the science behind what makes the data high quality and useful to train our models. We also develop and evaluate quantitative metrics that measure the success and impact of our data and training interventions.

Beyond execution, we explores new paradigms for human-ai interaction and scalable oversight, experimenting with how humans can best supervise, guide, and collaborate with models. It’s interdisciplinary work that blends research, data operations, and technical implementation to advance the frontier of aligned, human-centered AI systems.

This role blends fundamental research and practical engineering, as we do not distinguish between the two roles internally. You will be expected to write high-performance code and read technical reports. It’s an excellent fit for someone who enjoys both deep theoretical exploration and hands-on experimentation, and who wants to shape the foundations of how AI learns.

Note: This is an "evergreen role" that we keep open on an on-going basis to express interest in this research area. We receive many applications, and there may not always be an immediate role that aligns perfectly with your experience and skills. Still, we encourage you to apply. We continuously review applications and reach out to applicants as new opportunities open. You are welcome to reapply if you get more experience, but please avoid applying more than once every 6 months. You may also find that we put up postings for singular roles for separate, project or team specific needs. In those cases, you're welcome to apply directly in addition to an evergreen role.

What You’ll Do

• Design and execute data collection and synthesis strategies for post-training by combining human feedback, preference data, and synthetic examples to guide model behavior.
• Develop pipelines and frameworks for scalable, high-quality human labeling, model-assisted labeling, and synthetic data generation.
• Research and model human preferences and behavior, creating data-driven methods to improve reasoning, truthfulness, and helpfulness.
• Iterate on evals: post-training involves a never-ending loop of defining a set of evaluations, optimizing them, and then realizing your existing evals don’t capture what matters. You’ll be responsible for both making numbers go up, and making sure the numbers are meaningful.
• Design and evaluate metrics and benchmarks that measure data quality, alignment, and the real-world impact of post-training interventions.
• Scale and explore: post-training will involve a combination of scaling the existing methodologies and developing new ones.
• Publish and present research that moves the entire community forward. Share code, datasets, and insights that accelerate progress across industry and academia.

Skills and Qualifications

Minimum qualifications:

• Strong engineering skills, ability to contribute code and debug in complex codebases.
• Experience with data curation, human feedback, or synthetic data generation for large language models or similar systems.
• Ability to design, run, and interpret experiments with scientific rigor and clarity.
• Proficiency in Python and familiarity with at least one deep learning framework (e.g., PyTorch, TensorFlow, or JAX). Comfortable with debugging distributed training and writing code that scales.
• Bachelor’s degree or equivalent experience in Computer Science, Machine Learning, Physics, Mathematics, or a related discipline with strong theoretical and empirical grounding.
• Clarity in communication, an ability to explain complex technical concepts in writing.

Preferred qualifications — we encourage you to apply even if you don’t meet all preferred qualifications, but at least some:

• A strong grasp of probability, statistics, and ML fundamentals. You can look at experimental data and distinguish between real effects, noise, and bugs.
• Prior experience with RLHF, RLAIF, preference modeling, or reward learning for large models.
• Experience managing or analyzing human data collection campaigns or large-scale annotation workflows.
• Research or engineering contributions in alignment, data-centric AI, or human-AI collaboration.
• Familiarity with synthetic data pipelines, active learning, or model-assisted labeling
• PhD in Computer Science, Machine Learning, Physics, Mathematics, or a related discipline with strong theoretical and empirical grounding; or, equivalent industry research experience.

Logistics

• Location: This role is based in San Francisco, California. 
• Compensation: Depending on background, skills and experience, the expected annual salary range for this position is $350,000 - $475,000 USD.
• Visa sponsorship: We sponsor visas. While we can't guarantee success for every candidate or role, if you're the right fit, we're committed to working through the visa process together.
• Benefits: Thinking Machines offers generous health, dental, and vision benefits, unlimited PTO, paid parental leave, and relocation support as needed.

About the Role

The role of post-training researchers sits at the core of our roadmap. This is the critical bridge between raw model intelligence and a system that is actually useful, safe, and collaborative for humans.

This role blends fundamental research and practical engineering, as we do not distinguish between the two roles internally. You will be expected to write high-performance code and read technical reports. It’s an excellent fit for someone who enjoys both deep theoretical exploration and hands-on experimentation, and who wants to shape the foundations of how AI learns.

Note: This is an "evergreen role" that we keep open on an on-going basis to express interest in this research area. We receive many applications, and there may not always be an immediate role that aligns perfectly with your experience and skills. Still, we encourage you to apply. We continuously review applications and reach out to applicants as new opportunities open. You are welcome to reapply if you get more experience, but please avoid applying more than once every 6 months. You may also find that we put up postings for singular roles for separate, project or team specific needs. In those cases, you're welcome to apply directly in addition to an evergreen role.

What You’ll Do

• Develop and tune the recipe: iterate on post-training recipes, consisting of a collection of datasets, training stages, and hyperparameters. Measure how recipe choices affect various metrics.
• Iterate on evals: post-training involves a never-ending loop of defining a set of evaluations, optimizing them, and then realizing your existing evals don’t capture what matters. You’ll be responsible for both making numbers go up, and making sure the numbers are meaningful.
• Debug and understand: while tuning the details of a training configuration, we often observe results that don’t quite make sense. You’ll be responsible for both getting things to work, and developing a deeper understanding, which we can bring to the next problem.
• Scale and explore: post-training will involve a combination of scaling the existing methodologies and developing new ones. We’ll want to both measure how performance metrics scale with dataset size, and explore using a completely different kind of training dataset.
• Publish and present research that moves the entire community forward. Share code, datasets, and insights that accelerate progress across industry and academia.

Skills and Qualifications

Minimum qualifications:

• Proficiency in Python and familiarity with at least one deep learning framework (e.g., PyTorch, TensorFlow, or JAX). Comfortable with debugging distributed training and writing code that scales.
• Bachelor’s degree or equivalent experience in Computer Science, Machine Learning, Physics, Mathematics, or a related discipline with strong theoretical and empirical grounding.
• Clarity in communication, an ability to explain complex technical concepts in writing.

Preferred qualifications — we encourage you to apply even if you don’t meet all preferred qualifications, but at least some:

• A strong grasp of probability, statistics, and ML fundamentals. You can look at experimental data and distinguish between real effects, noise, and bugs.
• Prior experience with RLHF, RLAIF, preference modeling, or reward learning for large models.
• Experience managing or analyzing human data collection campaigns or large-scale annotation workflows.
• Research or engineering contributions in alignment, data-centric AI, or human-AI collaboration.
• PhD in Computer Science, Machine Learning, Physics, Mathematics, or a related discipline with strong theoretical and empirical grounding; or, equivalent industry research experience.

Logistics

• Location: This role is based in San Francisco, California. 
• Compensation: Depending on background, skills and experience, the expected annual salary range for this position is $350,000 - $475,000 USD.
• Visa sponsorship: We sponsor visas. While we can't guarantee success for every candidate or role, if you're the right fit, we're committed to working through the visa process together.
• Benefits: Thinking Machines offers generous health, dental, and vision benefits, unlimited PTO, paid parental leave, and relocation support as needed.

About the Role

Thinking Machines builds multimodal-first. For us, there is no separate multimodal work. It’s at the core of everything we do, from the scientific goals we’re setting to the infrastructure we’re building. We’re looking for researchers to advance the frontier of audio capabilities. You’ll explore how audio models enable more natural and efficient communication/collaboration, preserving more information and capturing user intent.

This is a highly collaborative role. You’ll work closely across pre-training, post-training, and product with world-class researchers, infrastructure engineers, and designers. This is an opportunity to shape the fundamental capabilities of AI systems that millions of people will use.

This role blends fundamental research and practical engineering, as we do not distinguish between the two roles internally. You will be expected to write high-performance code and read technical reports. It’s an excellent fit for someone who enjoys both deep theoretical exploration and hands-on experimentation, and who wants to shape the foundations of how AI learns.

Note: This is an "evergreen role" that we keep open on an on-going basis to express interest in this research area. We receive many applications, and there may not always be an immediate role that aligns perfectly with your experience and skills. Still, we encourage you to apply. We continuously review applications and reach out to applicants as new opportunities open. You are welcome to reapply if you get more experience, but please avoid applying more than once every 6 months. You may also find that we put up postings for singular roles for separate, project or team specific needs. In those cases, you're welcome to apply directly in addition to an evergreen role.

What You’ll Do

• Own research projects on audio training, low-latency inference and conversational responsiveness.
• Design and train large-scale models that natively support audio input and output.
• Investigate scaling behavior such as how data, model size, and compute affect capability and efficiency.
• Build and maintain audio data pipelines, including preprocessing, filtering, segmentation, and alignment for training and evaluation.
• Collaborate with data and infrastructure teams to scale audio training efficiently across distributed systems.
• Publish and present research that moves the entire community forward. Share code, datasets, and insights that accelerate progress across industry and academia.

Skills and Qualifications

Minimum qualifications:

• Ability to design, run, and analyze experiments thoughtfully, with demonstrated research judgment and empirical rigor.
• Understanding of machine learning fundamentals, large-scale training, and distributed compute environments.
• Proficiency in Python and familiarity with at least one deep learning framework (e.g., PyTorch, TensorFlow, or JAX). Comfortable with debugging distributed training and writing code that scales.
• Bachelor’s degree or equivalent experience in Computer Science, Machine Learning, Physics, Mathematics, or a related discipline with strong theoretical and empirical grounding.
• Clarity in communication, an ability to explain complex technical concepts in writing.

Preferred qualifications — we encourage you to apply even if you don’t meet all preferred qualifications, but at least some:

• A strong grasp of probability, statistics, and ML fundamentals. You can look at experimental data and distinguish between real effects, noise, and bugs.
• Experience with real-time inference, streaming architectures, or optimization for low latency.
• Prior experience training or evaluating large-scale audio or multimodal models.
• Publications, releases, or open-source projects related to speech, audio, voice, or similar areas.
• Demonstrated experience in audio or speech modeling, including ASR, TTS, or self-supervised audio learning.
• PhD in Computer Science, Machine Learning, Physics, Mathematics, or a related discipline with strong theoretical and empirical grounding; or, equivalent industry research experience.

Logistics

• Location: This role is based in San Francisco, California. 
• Compensation: Depending on background, skills and experience, the expected annual salary range for this position is $350,000 - $475,000 USD.
• Visa sponsorship: We sponsor visas. While we can't guarantee success for every candidate or role, if you're the right fit, we're committed to working through the visa process together.
• Benefits: Thinking Machines offers generous health, dental, and vision benefits, unlimited PTO, paid parental leave, and relocation support as needed.