Brandon Abranovic, Ph.D.
Research Scientist and Machine Learning Engineer
Company: brandon[AT]polymathmade[DOT]com
Personal: brandon[AT]abranovic[DOT]com
Voice: transcribe and send email
While I am always happy to chat, if you are an investor, my co-founder Katherine (katherine[AT]polymathmade[DOT]com) is generally the best person to contact for investment related inquiries.
I lead the development of Design Engine. Design Engine is a tool that brings AI-native capabilities — like autocomplete, prompt-based editing, and intent validation — into the heart of the CAD workflow. By enabling intelligent automation, we can help teams iterate faster and build more than they thought possible.
Worked on machine learning systems at Relativity Space, building out computer vision and process monitoring tools to help make 3D printed rockets possible. Got to apply cutting-edge ML to solve some really challenging manufacturing problems! 🚀
I’m drawn to problems at the intersection of physical systems and intelligence—especially where traditional engineering meets data-driven reasoning. The problems that interest me most are the ones where structure exists, but isn’t easily formalized, where solutions require both mathematical rigor and practical intuition. My work is often about compressing human expertise into scalable systems—whether that means using data-driven models to capture institutional engineering expertise accumulated over years of practice, or designing tools that close the feedback loop between design, simulation, and fabrication.
I’m especially interested in how we can abstract and encode physical knowledge into learning systems. I care about making machine learning not just accurate, but useful: interpretable, and deployable in real-world engineering workflows. These days, I’m focused on building systems that make engineering teams faster, more creative, and less bogged down by repetition.