Founding ML/Research Engineer
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
Company: Early-stage AI startup
Industry: AI / Machine Learning / Deep Tech
Location: San Francisco, California
Employment Type: Full-time
Work Arrangement: In-office
Salary: $150,000–$250,000
Visa Sponsorship: Limited
Experience: 3–6 years
Company Size: 1–10 employees
About the Company
An early-stage AI startup developing an engineering assessment platform for deep-tech companies. The platform evaluates engineers through realistic technical tasks, helping organisations identify suitable candidates and match engineering talent to complex projects.
The company works with businesses across space and defence, advanced manufacturing, robotics, and related technical sectors. Its product is already in production, and the company has recently secured pre-seed funding to support its growth.
About the Role
The company is hiring a Founding ML/Research Engineer to turn expert-calibrated engineering assessments and real hardware-engineering workflows into datasets, environments, and evaluation systems for training advanced AI models.
The role focuses on reinforcement learning (RL), agentic CAD and ECAD tasks, evaluation frameworks, reward signals, data pipelines, and machine learning infrastructure. You will work closely with mechanical and electrical engineers to define correct outcomes and collaborate with AI research teams using these environments for model post-training.
This is a hands-on machine learning engineering and research role, not an AI applications position focused primarily on retrieval-augmented generation (RAG) or prompt engineering.
Responsibilities
Design and implement reinforcement learning environments and evaluations for AI agents performing CAD and ECAD tasks.
Build grading systems, reward signals, and verification frameworks to assess engineering work reliably.
Transform engineering assessment data into datasets and environments suitable for AI model training and post-training.
Develop data pipelines aligned with the company's assessment workflows and research requirements.
Automate processes such as assessment scoring and question generation.
Collaborate with mechanical and electrical engineers to establish technical correctness and evaluation standards.
Work with AI research teams to support model training and post-training workflows.
Run experiments, analyse results, and contribute to research and product direction.
Identify opportunities to improve scalability, efficiency, and the quality of AI evaluations.
Requirements
Approximately 3–6 years of relevant professional experience in machine learning or AI engineering.
Strong grounding in deep-tech engineering through ML work at a hardware-focused company, close collaboration with electrical or mechanical engineers, or an educational background in electrical or mechanical engineering.
Experience with CAD, ECAD, 3D vision, or other hardware-adjacent machine learning applications.
Practical experience building, providing, or using reinforcement learning environments for model training or post-training.
Strong experience in machine learning evaluations, grading systems, data pipelines, or ML infrastructure.
Familiarity with reinforcement learning, evaluation frameworks, and LLM-as-a-judge approaches.
Hands-on engineering skills and the ability to implement systems rather than operate only at a conceptual level.
High initiative, strong problem-solving ability, and comfort working in an ambiguous environment.
Interview Process
Initial screening
Technical interview
Take-home assignment
Final round
Offer
🌍 Remote from anywhere in the globe
Workplace, Culture & Diversity
Corporate culture is essential to allow the organization to differentiate itself. In terms of image on the one hand, it has strengths both internally and externally to consumers. It is indeed a source of cohesion and motivation of employees and it limits conflicts. With customers, it conveys a positive image and can develop a feeling of closeness to the company or even become a criterion of choice.
About Fractal Talent
We Are Fractal Talent, a global community of freelance and fractional recruiters. We provide recruiting expertise to our clients as a virtual extension of their teams, deployed on an embedded, project or per role basis, and matched to client need. Top tier quality, unrivalled flexibility.
If you're a recruiter, come join a community that acts as your outsourced sales function. We align your skills to our clients - with the aim to enrich quality recruiters and enable a far greater flexibility in how you manage your career.