AI for Engineering Knowledge Management
How Leo AI’s Large Mechanical Model (LMM) empowers mechanical engineers to search CAD parts and generate assemblies.
·
⏱
2 min read
Dr. Maor Farid
Maor Farid is the Co-Founder and CEO of Leo AI, the first AI platform purpose-built for mechanical engineers. He holds a PhD in Mechanical Engineering and completed postdoctoral research at MIT as a Fulbright fellow. A Forbes 30 Under 30 honoree and former AI researcher and Mechanical Engineer in an elite military intelligence, Maor leads Leo AI's mission to transform how engineering teams design better products faster.
BOTTOM LINE
The gap between language models and engineering models is geometric, not linguistic. LLMs understand words about geometry — LMMs understand geometry itself. That difference determines whether an AI can be trusted to reason about physical systems.
For any engineering task where the output becomes a physical thing, that distinction is not academic.
The impact is quantifiable. Teams using Leo report 8.3+ hours saved per engineer per week. Design errors drop by 34%. And organizations see 211% faster time-to-market when their entire team has access to the same knowledge base instead of working in silos.
Yann LeCun says LLMs will never reach human-level understanding with text alone. Discover why geometry breaks GPT — and how Leo AI’s Large Mechanical Model (LMM) empowers mechanical engineers to search CAD parts and generate assemblies.
Yann LeCun, one of the godfathers of modern AI, recently said:
“Large language models will never get to human level by just training on text. We’ve got to train models to understand the real world.”
For mechanical engineers, this statement feels like common sense.
It doesn’t matter how massive GPT becomes,
how many billions of dollars are invested in OpenAI,
or how much data it consumes from the internet —
it will never match the skills of a first-year mechanical engineering student at a community college when it comes to something as basic as matching a bolt to a hole.
Why LLMs Can’t Think Like Engineers
The reason is simple: GPT is a language model. And language can’t describe geometry.
Ask any engineer to explain a mechanical concept, and within a minute, they’ll grab a marker and sketch it on a whiteboard (badly, usually). We think in shapes, constraints, and spatial relationships — not just words.
IN PRACTICE
What Engineers Are Saying
"Unlike general AI, Leo uses a Large Mechanical Model trained on 1M+ technical sources — standards, textbooks, datasheets. It also provides citations, so we don't have to guess whether a material property or tolerance is correct."
— Dorian G., AI Engineer
From Words to Parts: Introducing the Large Mechanical Model (LMM)
That’s why at Leo AI, we built the first Large Mechanical Model (LMM).
Unlike LLMs that use words as tokens, the LMM uses mechanical parts as tokens. It’s designed to understand CAD — not just read descriptions of CAD.
This allows engineers to:
Search CAD parts across internal PLM systems and online catalogs using free language.
Understand geometric constraints and relationships between parts.
Soon: Generate assemblies and complete product designs directly from a description.
Why We Built It
We’re not building a new foundational model because we want to look cool.
We don’t care about writing patents for the sake of press coverage or publishing papers to collect citations.
We built it because we had no choice.
For engineers to search and design effectively with AI, the model must think in geometry — something LLMs will never do.
FAQ
Stop Wasting Hours on Manual CAD Search
Leo AI turns your existing vault into a searchable knowledge base.
Leo AI connects to your PDM and makes every part findable by description in under 10 seconds. <a href="/onboarding">Try Leo Today</a>
Schedule a Demo →
#1 New AI Software Globally - G2 2026
Enterprise-grade security
Trusted by world-class engineering teams
Stop Wasting Hours on Manual CAD Search
Leo AI turns your existing vault into a searchable knowledge base.
Leo AI connects to your PDM and makes every part findable by description in under 10 seconds. <a href="/onboarding">Try Leo Today</a>
Schedule a Demo →
#1 New AI Software Globally - G2 2026
Enterprise-grade security
Trusted by world-class engineering teams