Location: San Mateo, CA (In-Person, Bay Area)
About PrimepointPrimepoint is building construction intelligence for the physical world. We transform massive, static construction datasets into an interactive AI system that surfaces conflicts and reduces risk on $100M–$3B+ building projects.
We go beyond LLM wrappers and focus on multimodal reasoning across diagrams, text, and structured data.
Founders:
Founded Facebook’s first Computer Vision team
Helped build Facebook AI Research
Built and sold a neural video compression startup to Apple
Team:
~$10M seed
~3 years runway
Paying customers
We are hiring an Applied ML / Computer Vision Research Engineer with a strong academic background and product mindset.
You will:
Develop models for understanding technical drawings and specifications
Work on multimodal reasoning across diagrams, text, and structured data
Translate research into production systems within ~1 year horizons
Present prior research and apply it to real-world constraints
Collaborate closely with founders
This is applied research. Everything must impact customers.
What We’re Looking ForMust-Have
PhD in Machine Learning, Computer Vision, or related field
Publications in top-tier venues (CVPR, ICCV, ECCV, ICML preferred)
Applied orientation (not purely theoretical)
Strong engineering capability
Based in Bay Area, in-person
Ideal, but not required
Document understanding / diagram analysis
Vision-language / multimodal experience
Knowledge graphs / structured reasoning
Production ML experience
Skills Required
- PhD in Machine Learning, Computer Vision, or related field
- Publications in top-tier venues (CVPR, ICCV, ECCV, ICML)
- Applied orientation (not purely theoretical)
- Strong engineering capability
- Based in Bay Area, in-person
- Document understanding / diagram analysis
- Vision-language / multimodal experience
- Knowledge graphs / structured reasoning
- Production ML experience
What We Do
Primepoint is a construction intelligence platform that leverages AI to understand construction drawings and project documents. By reading linework and connecting documents like specifications and RFIs, the platform automates manual coordination tasks. This allows project teams to identify risks earlier, reduce rework, and improve overall productivity throughout the construction lifecycle, from preconstruction to project closeout.








