Top Tech Jobs & Startup Jobs in NYC, NY

Conduct original research to evaluate, train, and improve frontier AI systems. Responsibilities include developing STEM benchmarks, synthetic data methods, reliability and hallucination evaluations, verification techniques, and agentic science workflows. The role involves identifying research gaps, designing rigorous experiments, building evaluation pipelines, publishing findings, and proposing new research directions across reasoning, AI evaluation, data generation, and AI-for-science.

Conduct original research to evaluate, train, and improve frontier AI systems. Design benchmarks, synthetic-data methods, reliability and hallucination evaluations, verification techniques, and agentic science workflows. Identify research gaps, formulate hypotheses, build experiments and evaluation pipelines, publish findings, and develop new research directions across STEM and AI. The role requires strong Python skills, quantitative reasoning, technical writing, and expertise with modern LLMs.

Conduct frontier AI research and build research-grade datasets, prototypes, tooling, reinforcement learning environments, benchmarks, and evaluation frameworks. Investigate synthetic and agentic data generation, post-training, model understanding, and model capabilities. Design rigorous experiments, analyze results, and translate findings into scalable products and AI systems while collaborating across research, engineering, product, and operations teams.

Investigate fraud and insider-threat cases from initial signal through resolution. Gather and analyze logs, transactions, behavioral signals, and other evidence; triage alerts; reconstruct events; document findings objectively; maintain evidence chain of custody; and collaborate with Detection & Response, Legal, and HR. The role also contributes to investigation playbooks, detection improvements, and fraud and insider-risk prevention.

Own the end-to-end development and delivery of reinforcement learning environments for frontier AI labs and enterprise clients. Lead environment design, task generation, reward and verifier development, data quality, evaluation, and post-training experiments. Build and manage cross-functional engineering, research, domain expert, QA, and data operations teams. Establish technical processes, schemas, APIs, validation scripts, and review systems while managing customer relationships, solutioning, delivery, and project growth.

Owns end-to-end development of datasets, reinforcement learning environments, and evaluations for coding agents and software engineering models. Leads engineers, researchers, QA professionals, and data creators; establishes quality processes, pipelines, validation, and review workflows. Collaborates directly with frontier AI labs, fine-tunes models, builds benchmarks, develops automation and internal tools, and supports customer-facing enablement. The role combines hands-on technical leadership, engineering management, operational execution, research, and industry thought leadership.

The Staff Research Engineer - Code will lead the creation of datasets and reinforcement learning environments for coding agents, collaborating with AI labs to influence model improvements and operational excellence, while managing a technical team.