Applied Researcher

River AI Inc.; Palo Alto (CA); On-site; USD 200,000 - 420,000; Full time

Benefits offered by this job

Health insurance

Dental and vision insurance

Unlimited PTO

Relocation assistance

Visa sponsorship

Job summary

River AI Inc. is seeking exceptional applied researchers to work directly with customers across industries to train and improve their own models.

You will own the applied research lifecycle from problem definition to data preparation, experiment design, training, and evaluation, collaborating with River's research and infrastructure teams to apply modern deep learning techniques to practical customer problems.

Qualifications

• BS, MS, or Ph.D. in Computer Science, Machine Learning, Mathematics, or equivalent practical industry experience.
• Strong understanding of modern deep learning architectures and practical experience training, fine-tuning, and evaluating language or multimodal models.
• Extensive hands-on experience with PyTorch or JAX, with a track record of writing clean, scalable code for model training and experimentation.
• Proven ability to turn open-ended problems and imperfect real-world data into working models, using rigorous experiments and evaluation to guide decisions.
• Strong communication and collaboration skills, with the ability to work directly with customers, learn new domains, and explain technical tradeoffs to technical and non-technical audiences.
• A highly collaborative mindset, ownership of outcomes, and a bias for action in a fast-paced environment.

Responsibilities

• Partner directly with customers across industries to understand workflows, identify high-value modeling problems, and define success criteria.
• Develop data preparation and curation pipelines with customer teams, turning domain-specific data and expert feedback into training datasets and representative evaluations.
• Design and run training experiments using techniques such as continued pre-training, supervised fine-tuning, reinforcement learning, and parameter-efficient fine-tuning (PEFT), selecting approaches that fit each customer's goals, data, and compute budget.
• Build evaluation suites, analyze model failures, and iterate on data and training methods to improve quality, reliability, and performance on real customer tasks.
• Work with customer engineers and River's infrastructure team to scale training runs, resolve bottlenecks, and support deployment and ongoing model improvement.
• Turn lessons from customer engagements into reusable training recipes, evaluation tools, and improvements to River's platform.

Skills

Deep learning architectures

PyTorch

JAX

Experimental design

Communication

Collaboration

Education

BS/MS/PhD in CS/ML/Math

Tools

PyTorch

JAX

Job description

At River, our mission is to create personal AI owned and shaped by each individual. To achieve this, we are rewriting the entire stack from scratch: personal hardware for local inference, bespoke training infrastructure, next-generation UIs, and frontier deep learning research.

Who we are

We are scientists, engineers, and builders from the industry's top tech companies and AI labs. We bring a proven track record of scaling consumer systems for hundreds of millions of users and architecting the pre-training infrastructure behind today's frontier models.

About the Role

We are looking for exceptional applied researchers to work directly with customers across industries to train and improve their own models. Your goal is to translate real-world problems, domain expertise, and customer data into models that deliver measurable value in their workflows.

You will own the applied research lifecycle, from defining the problem and preparing data to designing experiments, training models, and evaluating results. Working closely with customer teams and River's research and infrastructure teams, you will apply and adapt modern deep learning techniques to solve practical problems and help customers bring their models into production.

What You’ll Do

• Partner directly with customers across industries to understand their workflows, identify high-value modeling problems, and define success criteria.
• Develop data preparation and curation pipelines with customer teams, turning domain-specific data and expert feedback into training datasets and representative evaluations.
• Design and run training experiments using techniques such as continued pre-training, supervised fine-tuning, reinforcement learning, and parameter-efficient fine-tuning (PEFT), selecting approaches that fit each customer's goals, data, and compute budget.
• Build evaluation suites, analyze model failures, and iterate on data and training methods to improve quality, reliability, and performance on real customer tasks.
• Work with customer engineers and River's infrastructure team to scale training runs, resolve bottlenecks, and support deployment and ongoing model improvement.
• Turn lessons from customer engagements into reusable training recipes, evaluation tools, and improvements to River's platform.

Skills & Qualifications

Minimum Qualifications:

• BS, MS, or Ph.D. in Computer Science, Machine Learning, Mathematics, or equivalent practical industry experience.
• Strong understanding of modern deep learning architectures and practical experience training, fine-tuning, and evaluating language or multimodal models.
• Extensive hands-on experience with PyTorch or JAX, with a track record of writing clean, scalable code for model training and experimentation.
• Proven ability to turn open-ended problems and imperfect real-world data into working models, using rigorous experiments and evaluation to guide decisions.
• Strong communication and collaboration skills, with the ability to work directly with customers, learn new domains, and explain technical tradeoffs to technical and non-technical audiences.
• A highly collaborative mindset, ownership of outcomes, and a bias for action in a fast-paced environment.

Preferred Qualifications: (We encourage you to apply even if you don't meet all of these)

• Experience delivering models for customer or industry use cases, with measurable improvements in model performance or business outcomes.
• Expertise in reinforcement learning, preference optimization, synthetic data generation, or adapting models to specialized domains.
• Experience with distributed training, experiment management, or deploying and monitoring models in production.
• Experience partnering with domain experts and working with proprietary or sensitive datasets under customer privacy and security requirements.
• Compensation: Depending on experience and skills the expected base pay is $200,000 - $420,000 USD per year, plus equity.
• Benefits: Comprehensive health, dental, and vision insurance; unlimited PTO; and relocation assistance as needed.
• Visa Sponsorship: We sponsor visas and are committed to supporting the process for the right candidate.