INDUSTRY · FOUNDRY

AI for foundries

2 weeks to 3 months from diagnosis to production

In a foundry, a defect found at finishing is value already lost. We design custom AI solutions that inspect your parts right out of the mold, anticipate furnace drift and stabilize your pouring processes.

Industry challenges in 2026

Foundries face heavy constraints: costly scrap, energy, scarce know-how and processes that are hard to stabilize. AI addresses each of these constraints without changing how you work.

A row of machines in a workshop
  • 01

    Defects caught too late.

    Porosity, shrinkage and surface defects often only show up at finishing, when the value is already sunk. AI vision spots them right out of the mold and triggers a correction before an entire batch is scrapped.

  • 02

    Energy-hungry melting.

    Melting accounts for most of your energy bill. AI optimizes furnace parameters at constant quality and flags abnormal overconsumption in real time.

  • 03

    Know-how that is getting scarce.

    Adjusting a melt or reading a defect relies on the experience of a few people. AI captures these decisions in models and assistants, so the know-how stays available as the team changes.

  • 04

    Processes that are hard to stabilize.

    Pouring temperature, alloy composition, sand quality: many parameters affect quality. AI tracks them continuously and suggests the settings that stabilize your conformity rate.

Roles in the industry

From melting to finishing, every role in the foundry can rely on AI to secure quality, reduce scrap and preserve scarce know-how.

A foundry worker in protective gear beside metal being poured into moulds
  • Melting operator.

    Oversees melting and pouring. AI tracks temperature and composition continuously to stabilize each melt and limit out-of-tolerance pours.

  • Molder.

    Prepares and assembles molds. AI vision checks mold condition and impression quality before pouring, to prevent serial surface defects.

  • Coremaker.

    Makes the cores. AI detects cracks and core defects before assembly, where a break often goes unnoticed until shakeout.

  • Quality manager.

    Oversees part inspection. AI spots porosity, shrinkage and surface defects right out of the mold, with full batch traceability.

  • Process engineering manager.

    Industrializes new parts. AI cross-references pouring parameters and quality results to set up a new part number faster.

  • Maintenance technician.

    Maintains furnaces and equipment. AI analyzes equipment signals to anticipate furnace drift before production stops.

Why StratIA

  • Sovereignty.

    Your data stays in France or on site. We rely on sovereign building blocks (OVH, Scaleway, Mistral) or on-premise deployment, with no dependency on a foreign cloud.

  • French team.

    A dedicated French team, from scoping to production. People who know your industrial context and stay available after deployment.

  • Defined ROI.

    Return on investment is defined from the scoping phase. We prioritize the most profitable use case and set measurable goals before writing a single line of code.

They trust us

  • Accessite logo, a StratIA client
  • Hello Prépa logo, a StratIA client
  • Bastide Manutention logo, a StratIA client
  • Paymetrust logo, a StratIA client
  • Comité National Olympique et Sportif Français logo, a StratIA client
  • Union des Marchands de Biens logo, a StratIA client
  • Ordre des Géomètres-Experts logo, a StratIA client
  • BTP Ingénierie logo, a StratIA client
  • EM Lyon Business School logo, a StratIA client
  • Logo of a StratIA client
  • Fandi logo, a StratIA client
  • Caratelli logo, a StratIA client
  • Maison Colombo logo, a StratIA client
  • Colibrimm logo, a StratIA client
  • Ligue Occitanie de Tennis logo, a StratIA client
  • Disney logo, a StratIA client

Expert perspectives

“Foundries hold considerable know-how, but too much value still ends up as scrap because defects are not caught early enough. AI does not replace the founder: it puts a camera and a model on the line to spot porosity or shrinkage right out of the mold, and closes the correction loop before finishing. That is where the return on investment is won, within a few months.”

  • Portrait of Yoan Hibert

    Yoan Hibert

    Co-founder & CEO

    Profile
  • Portrait of Baptiste Wieczorek

    Baptiste Wieczorek

    Co-founder & CTO

    Profile

Our method: learn, decide, deploy

You see the whole path from the start, and you only buy the first step. Each step has value on its own and prepares the next.

  1. Learn

    Bootcamp

    For whom: your teams first need to understand what AI changes in their work.

    One hands-on day, on site or remote, on your company's tool (Copilot, ChatGPT, Gemini, Claude), tailored to your industry and your team's level. Your teams build their first AI agents during the day and leave with the business use cases to tackle first.

    Price:
    €2,500 excl. VAT per group (up to 15 people), fundable through your OPCO (French training funds). Bootcamp+ version with follow-up of the same team over three months: €4,000 excl. VAT per group.
  2. Decide

    Bpifrance Diag Data IA

    For whom: you know AI matters, but not where to apply it first.

    A Bpifrance-accredited expert analyzes your organization and workflows, quantifies the value pools using your real numbers and gives you, for each one, the right path: do it yourselves, train a team, or deploy a tool.

    Price:
    €10,000 excl. VAT, 40% covered by Bpifrance, leaving €6,000 to pay (independent SME and mid-sized companies with 10 to 2,000 employees and at least €1M in revenue).

    Not eligible, or need something more focused? The AI audit, one day centered on your use cases, €3,000 excl. VAT.

  3. Deploy

    An AI agent on your workflow

    For whom: you want a result that runs, on the workflow that costs you the most (quotes, quality control, maintenance, documentation, reporting).

    We set up the agent on your existing tools (ERP, MES, CMMS), your teams approve its proposals, and it improves with every correction.

    Price:
    setup priced as a firm quote after the diag or the audit, then a subscription to keep it evolving.

An example

Molten metal being poured in a foundry

Example · Aluminum foundry

Detecting porosity before machining on cast parts

An aluminum foundry producing safety parts in medium runs was detecting porosity after machining, too late. We deployed vision inspection at the casting output, paired with a model trained on its recurring defects.

  • −21%

    scrap rate

  • 96%

    of defects detected before machining

  • 8 months

    return on investment

Estimated savings of €90,000 to €130,000 per year on material and machining hours avoided.

Discuss a similar project
A welder at work in a shower of blue sparks

Let's discuss your AI project

The questions business leaders ask us most

How does AI reduce scrap in a foundry?

By spotting defects right out of the mold, instead of discovering them at finishing or machining. The defect is isolated early, and pouring settings are corrected before an entire batch is scrapped.

Can we detect internal defects such as porosity?

Yes, if your parts already go through X-ray inspection. AI analyzes those images and spots porosity and shrinkage with the same rigor all day long, where the human eye tires.

Does AI handle small runs and many part numbers?

Yes. The models learn from few examples and update when a new part number arrives. That is essential in foundries, where runs are often short and varied.

Can AI help plan foundry production?

Yes. We have run a scheduling project in melting: AI cross-references orders, furnace constraints and priorities to suggest the best launch sequence, and recalculates it when a disruption hits.

How does AI help with furnaces?

On two fronts. It tracks furnace signals to spot drift before a breakdown, and it flags energy overconsumption at constant quality. On a station as critical as melting, that is where downtime costs the most.

How do we retain the know-how of our experienced founders?

By writing it down, where teams can find it. Adjusting a melt or reading a defect is captured in an assistant everyone can query at the workstation. The day a veteran leaves, their knowledge stays.