AI/ML Software Engineer | Nuclear AI
Remote — Continental US | $150k–$225k + Equity
We’re partnering with an innovative AI company building advanced technology for the nuclear energy sector, working alongside leading scientists and national laboratories to develop AI systems capable of understanding highly complex technical information.
They’re hiring an AI/ML Software Engineer to help build the models and systems behind a generative AI search platform used across the nuclear industry.
This is a hands-on engineering role focused on LLMs, RAG, information retrieval and document understanding — solving the challenge of making highly technical, domain-specific information accurately searchable and accessible.
What you'll work on:
- Build and improve production RAG systems for complex, domain-specific technical content
- Develop AI/ML models for document understanding, search, indexing and information retrieval
- Fine-tune and adapt language models for specialist nuclear terminology and content
- Build retrieval pipelines using embeddings, semantic search and modern information-retrieval techniques
- Develop safeguards and evaluation methods to reduce hallucinations, bias and unsupported outputs
- Build scalable, maintainable production ML systems using Python and PyTorch
- Work extensively with the Hugging Face ecosystem
- Process large-scale datasets and work with distributed computing environments
- Collaborate directly with scientists and nuclear-domain experts to improve model accuracy
- Contribute to the development of open-source AI tooling
- Work with world-class supercomputing resources for model training and experimentation
What we're looking for:
- 3+ years of professional AI/ML software engineering experience
- Strong Python engineering skills
- Hands-on production experience with PyTorch
- Strong understanding of NLP and Large Language Models
- Hands-on experience implementing and fine-tuning Retrieval-Augmented Generation (RAG) systems
- Strong understanding of information retrieval, particularly for search and document-heavy applications
- Experience with Hugging Face / Transformers
- Experience building robust, scalable and maintainable ML software
- Experience with large-scale data processing
- Strong communication and problem-solving skills
You don't need everything below, but we'd particularly like to hear from engineers with experience in: Hybrid/dense retrieval • BM25 • Semantic/vector search • Embeddings • Reranking • Document AI • Domain adaptation • LLM fine-tuning • Distributed training • Multi-GPU environments • HPC/supercomputing • LLM evaluation & hallucination mitigation • Open-source AI
Experience within nuclear, energy or another highly technical/regulated industry is valuable but not required.
This isn't about connecting an LLM API to a chatbot.
You'll be working on the underlying retrieval, model and document-understanding problems required to make AI reliably interpret decades of highly specialised nuclear information.
You'll have the opportunity to work with leading scientists, specialist domain experts and some of the world's most advanced computing infrastructure — while having significant ownership within a small engineering team.
Package:
$150,000–$225,000 base + equity
Fully remote — Continental US time zones
Health, dental & vision insurance
Professional development, conferences & training
Quarterly company meetups across the US
If you've built serious RAG, information retrieval or NLP/LLM systems in production and want to apply that experience to a technically difficult problem with real-world impact, we'd like to hear from you.
Apply to learn more about the company, team and technology.