Research, develop, and evaluate machine learning methods that improve product capabilities and customer outcomes

Design experiments to test new model architectures, training approaches, data strategies, and system designs

Adapt foundation models through fine-tuning, post-training, prompt optimization, retrieval, or other techniques

Develop evaluation frameworks and benchmarks for model quality, reliability, safety, and performance

Build prototypes and proofs of concept that demonstrate the potential of emerging AI techniques

Partner with AI/ML engineers and software engineers to translate successful experiments into production systems

Improve model accuracy, reasoning, latency, efficiency, robustness, and cost

Curate, generate, and evaluate datasets used for training, fine-tuning, and model assessment

Investigate model failures, edge cases, and unexpected behavior to identify opportunities for improvement

Stay current with relevant research and determine which advances can create practical product value

Communicate findings, tradeoffs, and technical recommendations to product, engineering, and executive stakeholders

Mentor other scientists and contribute to the company’s research culture, technical standards, and AI roadmap

Publish research, contribute to open-source projects, or represent the company within the broader technical community where appropriate

5+ years of experience in machine learning, artificial intelligence, applied research, or a related technical field

Strong foundation in deep learning, statistics, optimization, and experimental design

Experience developing or adapting models for real-world product applications

Expertise in one or more areas such as natural language processing, generative AI, computer vision, multimodal learning, reinforcement learning, recommendation systems, or speech

Proficiency in Python and modern machine learning frameworks such as PyTorch, TensorFlow, or JAX

Experience with model training, fine-tuning, post-training, evaluation, or inference

Ability to design rigorous experiments and draw sound conclusions from incomplete or ambiguous results

Track record of translating research concepts into prototypes, production systems, or measurable product improvements

Ability to collaborate closely with research, engineering, product, and domain experts

Strong written and verbal communication skills, including the ability to explain complex technical concepts clearly

Staff-level candidates may be expected to define research direction, lead cross-functional initiatives, and influence broader AI strategy

Advanced degree in computer science, machine learning, statistics, mathematics, or a related field may be preferred, although equivalent applied experience may be considered

Languages & Frameworks: Python, PyTorch, TensorFlow, JAX, Hugging Face, scikit-learn

Models & Techniques: Large language models, transformers, multimodal models, diffusion models, reinforcement learning, recommendation and ranking systems

Model Adaptation: Fine-tuning, reinforcement learning from feedback, preference optimization, distillation, synthetic data, prompt optimization

AI Systems: Retrieval-augmented generation, agents, tool use, structured generation, reasoning systems, model routing

Evaluation & Experimentation: Offline and online evaluation, human evaluation, benchmarking, red teaming, interpretability, model observability

Data & Infrastructure: Spark, Databricks, Snowflake, vector databases, distributed training, GPUs, Kubernetes

Models & Platforms: OpenAI, Anthropic, Google, Meta, open-source foundation models, and proprietary model architectures

Research, develop, and evaluate machine learning methods that improve product capabilities and customer outcomes

Design experiments to test new model architectures, training approaches, data strategies, and system designs

Adapt foundation models through fine-tuning, post-training, prompt optimization, retrieval, or other techniques

Develop evaluation frameworks and benchmarks for model quality, reliability, safety, and performance

Build prototypes and proofs of concept that demonstrate the potential of emerging AI techniques

Partner with AI/ML engineers and software engineers to translate successful experiments into production systems

Improve model accuracy, reasoning, latency, efficiency, robustness, and cost

Curate, generate, and evaluate datasets used for training, fine-tuning, and model assessment

Investigate model failures, edge cases, and unexpected behavior to identify opportunities for improvement

Stay current with relevant research and determine which advances can create practical product value

Communicate findings, tradeoffs, and technical recommendations to product, engineering, and executive stakeholders

Mentor other scientists and contribute to the company’s research culture, technical standards, and AI roadmap

Publish research, contribute to open-source projects, or represent the company within the broader technical community where appropriate

5+ years of experience in machine learning, artificial intelligence, applied research, or a related technical field

Strong foundation in deep learning, statistics, optimization, and experimental design

Experience developing or adapting models for real-world product applications

Expertise in one or more areas such as natural language processing, generative AI, computer vision, multimodal learning, reinforcement learning, recommendation systems, or speech

Proficiency in Python and modern machine learning frameworks such as PyTorch, TensorFlow, or JAX

Experience with model training, fine-tuning, post-training, evaluation, or inference

Ability to design rigorous experiments and draw sound conclusions from incomplete or ambiguous results

Track record of translating research concepts into prototypes, production systems, or measurable product improvements

Ability to collaborate closely with research, engineering, product, and domain experts

Strong written and verbal communication skills, including the ability to explain complex technical concepts clearly

Staff-level candidates may be expected to define research direction, lead cross-functional initiatives, and influence broader AI strategy

Advanced degree in computer science, machine learning, statistics, mathematics, or a related field may be preferred, although equivalent applied experience may be considered

Languages & Frameworks: Python, PyTorch, TensorFlow, JAX, Hugging Face, scikit-learn

Models & Techniques: Large language models, transformers, multimodal models, diffusion models, reinforcement learning, recommendation and ranking systems

Model Adaptation: Fine-tuning, reinforcement learning from feedback, preference optimization, distillation, synthetic data, prompt optimization

AI Systems: Retrieval-augmented generation, agents, tool use, structured generation, reasoning systems, model routing

Evaluation & Experimentation: Offline and online evaluation, human evaluation, benchmarking, red teaming, interpretability, model observability

Data & Infrastructure: Spark, Databricks, Snowflake, vector databases, distributed training, GPUs, Kubernetes

Models & Platforms: OpenAI, Anthropic, Google, Meta, open-source foundation models, and proprietary model architectures