Applied Scientist

Generalist • Somerville, MA

23h

About The Position

Applied Scientists at Generalist develop and improve frontier model capabilities, particularly through post-training recipes that enable robots to perform a wide range of dexterous tasks. Applications may include science lab automation, manufacturing, and warehouse logistics, as well as open-ended evaluations of model capabilities. The role owns the process of making manipulation capabilities robust, general, and reusable across applications, combining research, engineering, and operations.

This is an empirical, model science-first role. The Applied Scientist defines success criteria for capabilities, builds evaluations to measure them, runs experiments across model variants and data recipes, analyzes outcomes, and drives the post-training loop. The work uses post-training, data, task design, and rigorous experimentation to improve capabilities that can benefit the foundation model across industries.

Requirements

• Strong background in machine learning and/or robot learning; a PhD in a relevant area is given as an example.
• Comfort with large-scale data analysis, including slicing rollout logs, computing metrics, visualizing results, and investigating outcomes.
• Care about evaluation, benchmarking, and failure analysis, with diligence in repeated but changing task evaluations.
• Enjoy rapid iteration with real-world feedback and willingness to work hands-on with robots and physical task setups when needed.
• Motivation to make capabilities work for real applications.
• Ability to communicate findings clearly to research teams and partner-facing teams.

Responsibilities

• Own end-to-end improvement of target capabilities, from baseline performance to mastery.
• Design task suites and evaluation protocols that capture the needs of research, science partners, and customers and generalize beyond individual deployments.
• Coordinate data collection, train and fine-tune models, sweep data mixtures and configurations, and orchestrate large-scale evaluations.
• Conduct rigorous experimental analysis across model variants, including isolating variables, quantifying statistical confidence, and diagnosing failure modes from rollout data.
• Use the existing stack in new ways and work with pretraining and post-training teams to improve the model when a capability is fundamentally restricted.

What This Job Offers

• Job Type: Full-time
• Career Level: Mid Level
• Education Level: No Education Listed