Founding ML Engineer – Engineering Assessment AI

RNB Consultancy; San Francisco (CA); On-site; USD 150,000 - 250,000; Full time

Benefits offered by this job

Equity

Founding role

Job summary

Rnb Consultancy in San Francisco is seeking the founding ML engineer to build RL environments, graders and reward signals from real engineering work, shaping data for frontier labs. You’ll own post-training data pipelines and evaluation frameworks as the company grows.

You’ll contribute to research and product roadmap, work closely with hardware engineers, and operate on-site in San Francisco with a high-autonomy, high-impact role. This position offers equity and rapid growth potential.

Qualifications

• Real experience with RL environments built, sold or used for post-training
• A grounding in hard tech: ML at a hardware company or an EE/ME degree
• Knowledge of ML around CAD, ECAD, 3D or hardware data
• Minimum 3 years of experience
• Working from the San Francisco office, 9-9-6 schedule
• US citizen, green card holder, refugee or asylee, or TN case

Responsibilities

• Develop RL environments, graders and reward signals built from real engineering work
• Datasets shaped for post-training at frontier labs
• Evals, LLM-as-a-judge setups and the data pipelines behind them
• A real say in the research and product roadmap, in a role that grows with the company

Skills

RL environments

Post-training data pipelines

Reward signals

CAD/ECAD data

Education

EE/ME degree

Tools

CAD/ECAD software

Job description

You've built RL environments or graders, and hardware engineering doesn't scare you. Be the first ML hire on data nobody else has.

A pre-seed AI startup in San Francisco tests engineers on real engineering tasks for companies in space, defense, advanced manufacturing and robotics, so they can pick the right engineer for every project. It's already live with large deep-tech customers. As the founding ML engineer, you convert rubrics set by experts and recordings of real CAD and ECAD sessions into training environments, scoring models and reward signals, and package them as post-training data for frontier labs.

What you'll own

• RL environments, graders and reward signals built from real engineering work
• Datasets shaped for post-training at frontier labs
• Evals, LLM-as-a-judge setups and the data pipelines behind them
• A real say in the research and product roadmap, in a role that grows with the company

Must-haves

• Real experience with RL environments: building, selling or using them for post-training (RAG or prompt-engineering work alone won't fit)
• A grounding in hard tech: ML at a hardware company, close work with electrical or mechanical engineers, or an EE/ME degree
• Knowledge of ML around CAD, ECAD, 3D or other hardware data
• 3+ years of experience
• Working from the San Francisco office, 9-9-6 (moving to SF is expected if you live elsewhere)
• US citizen, green card holder, refugee or asylee, or a straightforward TN case

Bonus points

• Post-training experience at a scaling AI company or a frontier model team
• An EE or ME background that moved into ML
• A time you looked past your brief and simply did what the project needed

What's in it for you

• $150k–$250k base + 0.2–0.7% equity
• Founding ML seat in a new field, on proprietary data
• Real ML research, not an applications job

Good to know

• Full-time, on-site in San Francisco, 9-9-6
• No visa sponsorship beyond the options listed above
• Process: profile check → first call with the hiring lead → home assignment → final interview → offer