Member of Technical Staff, Research

The role

You will own a research agenda where research meets production by designing data and reward signals. You will architect research to advance model reasoning at scale and build long-horizon agentic tasks.

What you will do

• Architect and execute a research agenda to advance model reasoning at scale.
• Build long-horizon agentic tasks and graders.
• Diagnose failure modes by reading trajectories and creating data or reward signals.
• Partner with AI research teams on datasets and benchmarks for alignment and safety.
• Identify, scope, and run long projects for scaling data toward AGI.
• Work with engineering on data pipelines, internal tooling, and deep learning implementations.

Required qualifications

• Strong software engineering and deep Python.
• Comfortable owning a system end to end and debugging through layers you did not write.
• Built evaluations, RL environments, agent harnesses, verifiers, or execution sandboxes.
• Shipped research into production, particularly post-training, distillation, and high-stakes evaluation.
• Worked on large-scale agent systems including orchestration frameworks, tool APIs, distributed execution, observability, and logging.
• Ability to get to ground truth in an unfamiliar domain by building something and reading outputs.
• Ability to notice misspecified tasks, ambiguous prompts, and graders that can be gamed.
• Thoughtful perspective on the societal and safety impacts of deploying general-purpose AI systems.

Nice to have

• A PhD (CS, ML, NLP) and publications at NeurIPS, ICML, or ICLR.
• Work on public benchmarks or task suites, for example Terminal-Bench, SWE-bench-style harnesses, or HCAST.
• RLHF or RLAIF experience and a view on how training choices show up in model behavior.
• Experience running work through large expert or annotation operations.

Benefits

• Health
• dental
• vision
• flexible PTO

About the company

Abundant is an applied research lab focused on scaling reinforcement learning for safe and reliable agentic capabilities. The team includes researchers, roboticists, founders, and operators whose work includes the Waymo Driver.