Applied Research Scientist, Agents

About the role:

Lumafield’s AI is woven deeply throughout its products and platform. As an Applied Research Scientist, Agents, you will extend what is possible with Lumafield’s agentic tools. This involves working across the agent stack, from top level harness and eval development down to fine-tuning and pre-training of foundational encoders for spatial reasoning. You will get the chance to work with manufacturing and design data that is not available anywhere else.

What you'll do:

• Benchmark local LLMs, VLMs, and perception encoders.
• Fine tune and post-train models on domain-specific data and tasks.
• Develop harnesses and build real, deployed agents that our customers use.
• Develop harnesses and build real, deployed agents that our customers use.
• Develop evaluations that predict real world agent performance accurately.
• Partner with product and engineering to continually improve agents and agentic workflows.

About you:

• Are curious, detail-oriented, and a rigorous empiricist.
• People are often surprised by how fast you can build a prototype.
• Have a robust understanding of the fundamentals of LLMs, VLMs, and vision transformers.
• Are adept at using AI for leverage, not as a substitute for depth of understanding.
• Have strong python, data, and algorithms foundations.
• Are an excellent technical communicator.
• 5-10 years of experience in a relevant field.

Bonus:

• Experience in design, manufacturing, or hardware.
• Advanced degree in machine learning, optimization, or a relevant discipline.
• Experience with 2D or 3D data and associated perception tasks.