Summer AI Applied Research Internship — DC, Onsite

The Nuclear Company; Washington (District of Columbia); On-site; USD 34,000 - 39,000; Full time

Benefits offered by this job

Competitive compensation packages

401k with company match

Medical, dental, vision plans

Generous vacation policy, plus hugs

Job summary

The Nuclear Company is launching a 12-week Summer 2027 Data Science & Machine Learning Fellows program in Washington DC, on-site five days a week. Fellows will formulate problems, build simulations, and develop reinforcement learning or optimization models to inform fleet-scale nuclear decisions.

You will work with nuclear industry experts to deliver deployable solutions, with housing and relocation provided for non-DC residents and a focus on rigorous, reproducible research practices.

Qualifications

• Pursuing MS or PhD in a quantitative field with expected graduation 2027 or later.
• Production-quality Python and PyTorch
• Hands-on experience with RL, optimization, simulation, or forecasting

Responsibilities

• Formulate problems and translate operations into modeling problems (MDP or optimization).
• Build simulation environments and evaluate models.
• Develop ML models for schedule optimization, capital allocation, or anomaly detection.
• Collaborate with engineers to deploy and monitor models in production.

Skills

Python

PyTorch

Reinforcement learning

Optimization

Time-series forecasting

Experiment design

Version control

Education

MS or PhD in Computer Science, ML, OR, Applied Math, Economics, or Statistics

Tools

Git

Job description

The Nuclear Company is launching a 12-week Summer 2027 Data Science & Machine Learning Fellows program in Washington DC, on-site five days a week. Fellows will formulate problems, build simulations, and develop reinforcement learning or optimization models to inform fleet-scale nuclear decisions.

You will work with nuclear industry experts to deliver deployable solutions, with housing and relocation provided for non-DC residents and a focus on rigorous, reproducible research practices.