Research Engineer, Post-Training

Actively AI•San Francisco, CA

About The Position

Actively AI is building a platform that deploys Per-Account Agents across customers’ total addressable markets to research accounts, identify opportunities, and advance next steps. The Research Engineer, Post-Training will help accelerate models tailored to the go-to-market and sales domain. The product supports sales representatives and leaders throughout prospecting, deal execution, expansion, and retention, creating demand for higher-quality intelligence across these tasks. The role works directly with the founders to fine-tune model intelligence and advance performance in this domain. The engineer will iterate on datasets, training stages, and hyperparameters; build and refine evaluations that reflect meaningful outcomes; investigate unexpected training results; apply alignment methods to shape agent reasoning and behavior; and measure how performance scales with data and compute. The role calls for rigor, invention, and the ability to understand end-user tasks and domain context.

Requirements

  • At least 3 years of full-time experience.
  • A track record of advancing machine learning systems through post-training, alignment, or related methods, such as RLHF, RLAIF, preference modeling, or reward learning.
  • Strong fundamentals in probability, statistics, and machine learning theory.
  • Ability to analyze experimental data and distinguish real effects from noise and bugs.
  • Ability to understand end-user tasks and domain context, and develop empathy and intuition for judging quality.

Nice To Haves

  • Original contributions or experience working on similar problems in industry, preferably in startup environments.

Responsibilities

  • Develop and iterate on post-training recipes, including datasets, training stages, and hyperparameters.
  • Measure how training choices affect evaluations and production performance.
  • Design, optimize, and continually improve evaluations, ensuring that results are meaningful and reliable.
  • Investigate unexpected training results to understand their causes and carry insights forward to future work.
  • Apply and advance alignment techniques such as RLHF, RLAIF, and constitutional approaches to shape agent reasoning, actions, and collaboration with humans on long-horizon tasks.
  • Measure performance scaling with data and compute, and develop new methodologies when existing approaches reach their limits.
  • Rapidly understand the sales and go-to-market domain and user tasks to judge the quality of model outputs.

Benefits

  • Competitive early-stage equity
  • Health coverage
  • Dental coverage
  • Vision coverage
  • Unlimited PTO and recharge days
  • Catered lunch on Tuesday and Friday, and dinner every day
  • Fully stocked kitchen
  • Cutting-edge technology and tools
  • Annual off-sites and monthly events
  • Commuter benefits
  • Cozy office in NYC
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service