Research Engineer (Mountain View) - 17813

Somewhere•Mountain View, CA

About The Position

We are a fast-growing applied AI research lab focused on data and reinforcement learning (RL) (Reinenforcement Learning ) environment curation for training and evaluating advanced agents. Our work has powered state-of-the-art reasoning datasets , specialized frontier models , and multi-turn, tool-using agents trained via reinforcement learning . Our research and systems are actively used by multiple top-tier AI labs and enterprise teams. By combining deep research expertise with production-grade engineering, we are building the infrastructure layer that enables the next generation of agent training. We are uniquely positioned to capture significant market share in data-centric AI and RL environment design . We are seeking a Research Engineer to operate at the intersection of cutting-edge agent research and production-scale systems . In this role, you will work closely with frontier AI labs, enterprise customers, and internal research teams to design, build, and deploy high-quality RL environments at scale. You’ll translate research insights into robust, reproducible pipelines that directly impact how modern agents are trained and evaluated. This position is ideal for someone who enjoys: Reading and implementing new research Prototyping novel ideas quickly Scaling research artifacts into production systems Working directly with highly technical external partners You will have real ownership over systems that shape how next-generation agents learn.

Requirements

MS or PhD in Machine Learning, Computer Science, or a related field — or equivalent industry research experience
Demonstrated research contributions (publications, open-source work, or deployed research systems)
Strong understanding of reinforcement learning, agent training, or related fields
Ability to read, implement, and adapt ideas from recent research papers
Strong Python skills and experience with ML frameworks (e.g., PyTorch, JAX)
Experience building research infrastructure or production ML systems
Familiarity with cloud platforms (AWS, GCP) and distributed systems
Systematic approach to testing, validation, and quality assurance
Comfortable leveraging modern developer tools (e.g., AI-assisted coding workflows)
Excellent communication skills across research and engineering teams
Ability to translate research concepts into practical system requirements
Strong project scoping, prioritization, and execution skills
Comfortable presenting technical work to diverse, highly technical audiences
Understanding of what makes research artifacts valuable in real-world use
Experience shipping datasets, benchmarks, tools, or platforms used by others
Attention to documentation, usability, and long-term maintainability
Customer-oriented approach to solving technical problems

Nice To Haves

Hands-on experience training or evaluating RL agents
Background in data-centric AI, synthetic data, or dataset creation
Publications at top-tier ML conferences (NeurIPS, ICML, ICLR, etc.)
Prior experience as a Research Engineer or Applied Scientist
Contributions to widely used datasets, benchmarks, or evaluation suites

Responsibilities

Partner with frontier AI labs to understand agent training requirements and design custom RL environments
Stay current with advances in reinforcement learning, agent training, curriculum design, and evaluation
Prototype and validate new approaches to environment generation and data curation
Translate academic research into scalable engineering solutions
Build and maintain scalable pipelines for creating, validating, and deploying RL environments
Design systems that ensure data quality, diversity, and reproducibility
Implement automated QA and verification processes for environments
Develop evaluation frameworks to measure environment effectiveness and training outcomes
Work directly with enterprise customers to understand domain-specific agent challenges
Customize environment suites, benchmarks, and evaluation setups for different use cases
Provide technical guidance on best practices for agent training and evaluation
Present research findings and system capabilities to technical stakeholders
Scale research prototypes into reliable, production-ready systems
Establish reproducible workflows and strong engineering standards
Create documentation and tooling for internal teams and external users
Monitor, optimize, and evolve systems as environment production scales