Founding Engineer, Model Training and Research
- Salary
- $200K–$400K
- Moves you to
- United States
- Support
- Visa sponsorship
- Posted
75,218 relocation jobs, straight from company career pages
100% free · New jobs every minute
Show job descriptionHide job description
We’re supporting a client with the recruitment requirement outlined below. If your recruiting expertise and networks align with the role, please apply to express your interest.
Once approved to work on the opportunity, you’ll receive the full brief, including the client name, role details, and commercial terms.
Role Details
Location: New York, NY (San Francisco or Toronto may also be considered)
Industry: AI
Salary: $200,000–$400,000
Equity: 1%
Employment Type: Full-time
Work Arrangement: In-office, 5 days per week
Work Intensity: 9-9-5
Experience: Approximately 2–3 years
Visa / Sponsorship: To be confirmed
Company Size: 1–10
About the Company
A stealth, early-stage AI company building an AI-powered operating system for large institutions. Its goal is to help organizations execute complex work across teams and systems using reliable, long-running AI agents.
The company is developing open models that learn from human work patterns, alongside infrastructure for deploying agents and potentially standalone developer products.
The founding team is currently very small, and the company is hiring its first engineers across model training, infrastructure, and forward-deployed engineering.
About the Role
The company is looking for a founding engineer focused on model training and research to improve the capabilities of AI agents performing complex, long-horizon tasks.
This role bridges research, model training, evaluation, and practical engineering. The engineer will work with open models, design pre-training and post-training experiments, and develop ways to turn human work traces into useful training data.
The key priority is hands-on model training and evaluation that produces measurable, reproducible improvements in agent performance.
Responsibilities
Design and run pre-training and post-training experiments using open models.
Develop systems that transform human work traces into useful training datasets.
Explore representations of workflows, context, and decision-making that improve agent performance.
Build evaluations to measure model quality and reliability across long-running tasks.
Develop reproducible training and experimentation workflows.
Diagnose training failures and distinguish genuine improvements from misleading results.
Collaborate with infrastructure engineers to bring successful research approaches into deployed systems.
Own research questions from initial hypothesis through experimentation and practical implementation.
Requirements
Hands-on experience with open models and pre-training, post-training, or both.
Strong programming and engineering skills, demonstrated through research implementations, training systems, or reproducible experiments.
Understanding of training data, model optimization, evaluation, and experimental design.
Experience building training pipelines, preparing datasets, and evaluating model performance.
Ability to independently design experiments, investigate failures, and interpret results.
Ability to connect research improvements to practical performance on long-running agent tasks.
Comfortable working autonomously in an early-stage environment.
Ideal Candidate Profile
Approximately 2–3 years of relevant experience.
Background in applied AI, open-model development, or post-training is preferred.
Independent research experience is also relevant, provided it includes hands-on implementation and credible experimental results.
Has personally owned model-training experiments rather than only contributing to a large research team.
Can demonstrate rigorous evaluation practices and reproducible results.
Experience training models for agent tasks or learning from human behavior and workflow traces is a plus.
Familiarity with distributed training or training-efficiency improvements is valuable.
Open-source contributions, public experiments, research implementations, or entrepreneurial experience are additional positives.
Scrappy, entrepreneurial, and comfortable taking ownership with limited direction.
Academic credentials are flexible; demonstrated technical ability and practical research experience matter more than a specific educational background.
Location & Relocation Requirements
New York is the preferred location.
San Francisco and Toronto may also be considered.
Candidates based elsewhere must be willing to relocate.
The role requires working in person five days per week.
Candidates unwilling to relocate or work on-site are not aligned with the role.
Visa sponsorship and permitted work authorization statuses are still to be confirmed.
Benefits
Health insurance.
Dental insurance.
Vision insurance.
The broader benefits package is still being confirmed.
Interview Process
Founder Screen
Technical Interview – 90 minutes
In-Person Collaboration: Meeting in New York, San Francisco, or Toronto to assess collaboration, practical execution, and fit for a five-day in-office founding team.
Offer.
