HYLE INDUSTRIES

INTELLIGENCE,INDUSTRIALIZED.

Intelligence Engineering Matter

Hyle Industries is building intelligence-native infrastructure for engineering and manufacturing in the physical world.

The industrial economy
was not built for
machine intelligence

Machine intelligence is rapidly expanding what can be designed and engineered.

But its impact on the real economy will ultimately be determined by what it can make.

Today, the productive capacity of industry is not exposed in a form machine intelligence can use.

Factories, machines, suppliers, and production systems can do the work. Intelligence has no native way to understand that capacity and put it to use.

As intelligence scales, the constraint moves downstream.

01Reasoning

02Design

03Engineering

04Production

Last era

Next era

The constraint, drawn as a throat in a channel across four stations: reasoning, design, engineering, production. In the last era the throat sat over reasoning; in the next era it has moved to production.

Fig 01 /The constraint, by era

Production becomes the bottleneck.

The next great unlock is making industrial capability scale with intelligence.

Industry, rebuilt
for intelligence

The industrial systems of the last era were never designed for machine intelligence.

Hyle is rebuilding the layers intelligence must cross to reach the physical world.

Fig 02. A section through the stack. Intelligence descends through three layers, engineering, manufacturing and factories, in nine steps: reason, design and validate; plan, configure and route; make, inspect and measure. It travels as data on a hidden line, drawn dashed. Below manufacturing it crosses into matter and the line becomes an object line at twice the weight, over a hatched ground symbol; the steps below the crossing are marked by witness lines drawn to the line and stopping short of it, because an object line is not marked on. The measured result returns from measure along the foot of the drawing, carrying measured result, cycle time and failures, climbs the right of the sheet and arrives back at engineering. Below 900 pixels the same construction re-proportions, with the return running in a channel down the right.

Intelligence

Engineering

  1. 01Reason
  2. 02Design
  3. 03Validate

Intelligence that can work across geometry, physics, materials, simulation, and real design constraints.

Manufacturing

  1. 04Plan
  2. 05Configure
  3. 06Route

Manufacturing capability that can be understood, accessed, and acted on programmatically.

Intelligence crosses into matter

Factories

  1. 07Make
  2. 08Inspect
  3. 09Measure

Connected production infrastructure that exposes real machines, inspection, capacity, and physical outcomes.

Matter

  • Hidden line exists as data
  • Object line exists as matter
  • Return the measured result

Fig 02 /One part, three statesPlan · not to scale

Engineering informs production. Production returns reality to engineering.

Building for
intelligence-native industry

Hyle is building the systems that let machine intelligence engineer physical systems and access manufacturing directly.

03A/EngineeringIn development

Eidos

An engineering agent for the physical world

Eidos is an agent and engineering harness built to execute complex physical engineering work.

It works across CAD, geometry, simulation, mechanical design, electronics, materials, and manufacturing, using engineering software directly rather than operating inside a single tool.

Give Eidos an objective. It can reason through the problem, create and modify designs, run analyses and simulations, use engineering tools, and iterate toward a validated, manufacturable design.

Request early access

03B/ManufacturingEarly access

manufacturing.dev

Programmatic access to manufacturing

manufacturing.dev exposes manufacturing in a form software and agents can use.

Submit a part or assembly through an API, SDK, CLI, or MCP server. Get DFM feedback, compare processes and materials, receive price and lead time, and place the order.

The parts are manufactured, inspected, and shipped to you. Production and inspection data return in structured form.

API  /  SDK  /  CLI  /  MCP

Request early access

Production is
infrastructure

If production becomes the bottleneck, Hyle cannot stop at software.

Hyle operates factory infrastructure alongside the systems we build.

That gives us direct access to machines, process capability, inspection, and real production capacity, and lets us expose more of that capability to software and machine intelligence.

Hyle does not just build the software around industrial production. We participate directly in the industrial output it enables.

Factory 01 drawn from beside the aisle: the waterjet bed in the foreground with its bridge and a bracket toolpath under the cutting head, the mill's enclosure across the aisle showing two faces, the lathe and the inspection table beyond it, the printer bench and its tender on the left wall, portal frames and crane rails running to the end wall.

Fig 04 /Factory 01 · Southern Foundry · Marietta, GeorgiaTwo point · not to scale

The physical capacity has to be built too.

Factory 01/Southern Foundry/Marietta, Georgia

Factory 01 is Hyle's first production site.

We manufacture real customer parts here using machining, turning, waterjet cutting, additive manufacturing, and inspection.

Operating production directly lets us connect software to real manufacturing capacity, automate more of the path from quote to machine, and expose what the factory can actually do in a form intelligence can use.

Real machines. Real capacity.

01Waterjet
Flow Mach 500. 5 axis, 87 KSI, 13 × 6 ft bed.
02Mill
SYIL X9. 3 axis, 30 × 20 × 20 in, to 0.001 in.
03Lathe
Haas DS-30Y. Dual spindle, 13.75 × 32.5 in.
04Additive
Bambu H2S. Formlabs Form 4.
Materials
Aluminium, steel, stainless, tool steel, copper, brass, titanium, Inconel, carbon fibre, G10, stone, glass.
Quotes
Within 24 hours.

Visit Southern Foundry Get a part quoted

Make the ability
to make abundant

The intelligence era should expand more than our capacity to think. It should expand our capacity to build.

Machine intelligence is making engineering capability dramatically more abundant.

Historically, turning that capability into physical output required large concentrations of talent, capital, software, suppliers, factories, and coordination.

That kept industrial capability scarce.

Machine intelligence changes the leverage.

  1. Individual

    More engineering capability available to one person.

  2. Small team

    The ability to take on problems once reserved for much larger organizations.

  3. Intelligent system

    The ability to design, manufacture, test, learn, and iterate continuously.

The next step is making industrial capability scale with intelligence.

More people can attempt ambitious things.

  • More ideas can reach production.
  • More experiments can become real.
  • More ambitious things can be built.

The power to build becomes less concentrated.

The ability to make becomes abundant.

Work that has to
survive reality

At Hyle, software eventually meets matter.

  • A model becomes a design.
  • A design meets a machine.
  • A part gets measured.
  • The result either works or it doesn't.

We are building an interdisciplinary team across machine intelligence, geometry, engineering software, manufacturing, robotics, and industrial systems.

The problems do not divide cleanly. Neither do we.

Our researchers work beside mechanical engineers and machinists because the hardest problems cross the boundary between software and the physical world.

We want people who care about the whole result.

Open rolesMarietta, Georgia

Bring the thing you are proudest of having built.

View open roles