Exotec Skypod picking station: robots bring totes from the storage racks to an operator

    AI engineering system for regulated hardware

    Build and ship faster, without compromising rigor.

    Koddex helps engineering teams cut time to market for complex hardware products in regulated industries, with specialized AI agents that work alongside them on a fully traced source of truth.

    Backed by

    BreegaBpifranceEuraTechnologiesEuropean Union
    01

    Ship your next product months earlier, with AI agents in the team.

    Design, document, prove compliance: at each step, AI agents take on the tasks that fill an engineer's week, and your engineers keep every decision. Teams bring new products to market up to 30% faster.

    The V-model, accelerated
    Koddex in action, in 70 seconds
    1Specify
    Agents turn the customer specification into linked requirements
    Your engineers review and correct them.
    2Design
    Agents fill the BOM, compute mass and cost, and flag coverage gaps
    Your engineers make the design choices.
    3Release
    Agents check every change and prepare it for review
    Your engineers approve and freeze the baseline.

    03

    Cut repetitive engineering work in half. Your engineers keep every decision.

    Assign a task to an agent as you would to a colleague. It creates data, comments, moves the workflow forward and hands work to other agents, the way a coding agent writes code, opens pull requests and reviews them. Your engineers approve every result.

    −50%
    time spent on repetitive engineering tasks
    −30%
    time to market

    01

    They take the tasks you assign.

    An agent picks up the task, reads the parts, requirements and tests it needs, does the work on a branch and comments on what it found. You review the result like a colleague's.

    AI mechanical engineerService account · arm program

    Task

    The bracket moves to supplier B: +180 g. Update the mass budget, flag what is at risk and draft the change request.

    Plan

    1. Read the BOM and the bracket's links
    2. Run the mass roll-up in Python: arm 4.82 → 5.00 kg
    3. Flag REQ-112 (mass budget) and 3 tests to rerun
    4. Open branch auto/bracket-b and draft CR-118

    Output

    CR-118Bracket supplier change
    • Arm assembly mass4.82 → 5.00 kg
    • REQ-112 mass budgetmargin −2 %
    • Tests to rerun3
    • Evidenceattached
    Waiting for your approval
    Every action on the item's activitySimulated example

    AI verification engineer

    Agent settings · Koddex

    Configuring its skills
    • Structure a customer specification
    • Allocate requirements to the system
    • Write test skeletons from requirements
    • Run impact analysis on every change
    • Draft change requests for review
    Simulated run

    02

    You set their rules.

    You give each agent its skills, its access rights and its scope. On the V-model, the agent runs each phase, finds what fails, fixes it and repeats until every phase passes. The release is still signed by an engineer.

    03

    They work as a team.

    A pull request, a supplier file or a CAD revision starts the right agents automatically. Each one passes its result to the next, and the engineers concerned are notified to approve it.

    01

    A change comes in

    GitHub PR #42 · firmware
    Supplier PDF · bracket rev B
    CAD revision · arm v3

    02

    The agent finds what it affects

    REQ-112 mass budget at risk
    TEST fatigue to rerun
    Variant XL touched

    03

    Fixes proposed on branches

    auto/arm-rev-c
    auto/mass-budget-v4
    CR-118 drafted

    04

    Engineers notified, the decision stays human

    Reviewers notified
    Lead mechanical · approved
    Quality · reviewing
    Nothing merges into the baseline before sign-off Simulated example
    • AI engineering system assistant
    • AI requirements engineer
    • AI mechanical engineer
    • AI verification engineer
    • Your own agent
    • AI agents & MCP

    Koddex Forge

    Scale from one product to a full range without losing traceability.

    A secure, traced source of truth: every piece of data and every decision is versioned and verifiable, across millions of requirements, components and test results.

    Explore Koddex Forge
    54.0Ø 4403YOUR DATA MODELREQBOMTESTCADECO02ACCESS & TRACEABILITY01DATA AT SCALE

    01Millions of linked data points

    Requirements, components, test results and documents: Forge handles millions of linked items and stays fast as your programmes grow. Your CAD, PLM and ERP keep feeding it.

    02Granular access, full traceability

    Each user, partner and agent sees only the items its access rights allow, and every change is logged with who made it, person or agent.

    03Your own data model

    You define your object types, attributes and links in your own vocabulary. When your process changes, you update the model and your existing data follows.

    Koddex Forge works with any AI agent

    Koddex's own agents and the agents you connect through MCP, such as Claude or ChatGPT, work on the same data, within each user's permissions.

    AI agents & MCP

    Certify faster in the industries where proof is mandatory.

    Robotics

    From the first prototype to a fleet on three continents.

    Exotec runs its robot engineering on Koddex: requirements, BOM, compliance tests and every site, in one traced model.

    Koddex for robotics
    Exotec Skypod picking station: robots bring totes from the storage racks to an operator
    AMR · Rev DBaseline 5
    64
    BOM
    118
    Requirements
    37
    Tests
    Compliance tests
    • ISO 3691-4 · stoppingPassed
    • ISO 13849 · PL dVerified
    • EMC · EN 61000-6-2Passed
    Multi-site · Rev D
    • Site A · EuropeLive
    • Site B · North AmericaLive
    • Site C · AsiaRolling out
    1,000+ licences · Exotec

    Outsourced design offices

    One model per customer project, every deliverable proven.

    Answer specifications faster, run several customer baselines side by side, and hand over a compliance file your customer can audit.

    Koddex for design offices
    Two engineers reviewing drawings at a design office
    Tender response67 / 67
    REQ-019 · Take-off thrustCompliant
    REQ-022 · Mass ≤ 7.0 kg6.50 kg
    Customer projects
    • Client A · Baseline 3Locked
    • Client B · Baseline 2Delivered
    • Client C · Baseline 1In review
    Compliance matrix
    Exported
    Every requirement linked to its evidence
    Exotec

    Exotec gave 30% of engineering time back to the product.

    Exotec builds automated warehouse systems. Its requirements moved from PDFs into one engineering knowledge base in Koddex, shared with 20 partner organisations.

    1,000+
    users in Koddex
    1
    source of truth for product data (PDM)
    20
    partner organisations in the model

    Co-founder & CEO · Exotec

    −30%
    time to market for a new product
    −50%
    time spent on repetitive engineering tasks
    2 weeks
    1 day
    for a change impact analysis, instead of 2 weeks of meetings and spreadsheets
    3 weeks
    < 1 h
    to gather audit-ready evidence, instead of a 3-week effort
    −30 %, −50 % and < 1 h: ranges observed on guided deployments, not an audited average. 1 day: measured at our reference customer after a year in use. We walk through the calculation on your perimeter during the demo.

    A 20-min demo. Your first use case live in 1 month.

    In 20 minutes, we show Koddex on one of your use cases, with a product like yours and your vocabulary. Your first use case is then live in your own Koddex within a month.

    Frequently asked questions

    See all questions

    None of the three, and parts of all of them. Koddex is the AI engineering system for regulated hardware: it holds your product definition (requirements, BOM, tests, decisions, documents) as linked data that Koddex computes on: mass, cost, coverage and the impact of each change. It replaces the scattered spreadsheets between your tools rather than replacing your CAD, your ERP or your quality processes, and it connects to PLM, QMS, ERP and MES platforms via API.

    The first use case is typically live in production within 48 hours, and full onboarding takes about three weeks, not three months. Deployment follows a repeatable six-phase method (setup, learning, core concepts, data governance, use case implementation, wrap-up) with validation milestones, led by a dedicated Customer Success Manager. There is no multi-quarter integration project.

    Typically through your IT team's standard middleware layer, consuming the Koddex API. Data exchanges are fully traced: every integration action is logged with a per-integration audit trail, and access can be scoped precisely so an integration only reads or writes the data it needs. Your IT team keeps control of the exchange architecture.

    Exclusively in the European Union, in the AWS Europe (Ireland) region. No customer data is stored, processed or transferred outside the European Economic Area without your prior express written agreement, in compliance with GDPR Chapter V. If a non-EU authority requests access to your data, we commit contractually to notify you without delay, challenge legally baseless requests, and limit any disclosure to the strict minimum.