I AM HIRED

Неделя 310 ₽Месяц 630 ₽3 месяца 1 680 ₽Год 5 680 ₽

Founding Engineer, Model Training & Research — New York, United States

ML и AI, AI-инженерия · сегодня, 10:53 · 16.11.2026

Зарплата не указана

Откликнуться

Компания: ••••••••

Локация: New York, United States

Формат: не указан

Занятость: Полная занятость

Грейд: Senior

Founding Engineer, Model Training & Research

Location: New York (San Francisco or Toronto may also be considered)

Working arrangement: Fully onsite, 5 days per week (9-9-5)

Compensation: $200,000–$400,000 base salary + 1% equity

Experience: 2+ years (strong independent research and practical projects count; academic credentials are flexible)

About the company

A stealth, seed-stage AI startup founded in 2026 by a repeat founder, building a growth operating system for large institutions. The company is developing reliable AI systems capable of carrying out complex, long-running workflows across teams and systems. Its approach combines autonomous AI agents with open models that learn from how people work. Currently a team of one, the company is looking for a Founding Engineer to help build its model training and research capabilities from the ground up.

The opportunity

As Founding Engineer, Model Training & Research, you will take ownership of the company's open-model training and research strategy. You will design and run pre-training and post-training experiments, transform human workflow traces into useful training data, and develop evaluations to improve the reliability of long-running AI agents.

Working directly with the founder, you will have significant autonomy over research questions, technical implementation and experimentation. You will be expected to take ideas from initial hypotheses through hands-on engineering, evaluation and ultimately deployed systems. This is an opportunity to combine practical machine learning engineering with applied AI research while helping shape the technical direction of an early-stage company.

What you'll be doing

• Design and execute pre-training and post-training experiments on open-weight models.

• Develop training approaches that enable AI models to learn from human workflows and real-world tasks.

• Turn human work traces and interaction data into high-quality training datasets.

• Build evaluation frameworks to measure the reliability and performance of long-running AI agents.

• Investigate failed experiments, identify underlying issues and improve training methodologies.

• Develop and implement research ideas end to end, including the engineering infrastructure required to support them.

• Evaluate training improvements through reproducible experiments and measurable results.

• Work directly with the founder on research priorities, technical architecture and product development.

What we're looking for

• 2+ years of relevant experience in machine learning engineering, research engineering or model training.

• Hands-on experience training open-weight or open-source models, with meaningful depth in pre-training, post-training or both.

• Experience with techniques such as fine-tuning, reinforcement learning, RLHF or other model optimisation approaches.

• Strong practical skills in implementing, running and evaluating machine learning experiments.

• Experience building or working with training pipelines, evaluation frameworks or distributed training infrastructure.

• Ability to independently own research projects from hypothesis and experimentation through implementation and evaluation.

• Strong analytical and problem-solving skills, with the ability to distinguish genuine improvements from misleading experimental results.

Independent research, open-source contributions, public model releases and practical projects are highly valued. Academic credentials and formal education are flexible. Demonstrated hands-on ability and individual ownership matter more than institutional background or years of experience.

What makes this opportunity interesting?

• Become a founding engineer at an ambitious, newly established AI startup.

• Work directly with an experienced repeat founder.

• Own the model training and research direction from day one.

• Work on open models, agent training and learning from real-world human workflows.

• Have the autonomy to turn research ideas into deployed systems.

• Play a central role in shaping the company's technology and future product.

Location and working arrangements

This is a fully onsite position, five days per week, following a 9-9-5 working schedule. New York is the preferred location, although exceptional candidates based in San Francisco or Toronto may also be considered. Candidates elsewhere must be willing to relocate to one of these locations.

Current work authorisation and potential visa sponsorship requirements will be discussed during the recruitment process. Visa sponsorship availability is yet to be confirmed.

Compensation and benefits

• Base salary of $200,000–$400,000.

• 1% equity.

• Founding-level ownership and influence over the company's research and technical direction.

• Direct collaboration with the founder.

• Opportunity to build the company's model training and research capabilities from the ground up.

Поиск по вакансиям

Название компании скрыто

Названия компаний и их продуктов мы не публикуем на сайте — они доступны в Telegram-боте I AM HIRED. Откройте эту вакансию в боте: увидите название, сможете откликнуться и настроить рассылку подходящих вакансий.

Открыть вакансию в боте