SPREAD AI is positioning as a series b horizontal AI infrastructure play, building foundational capabilities around knowledge graphs.
As agentic architectures emerge as the dominant build pattern, SPREAD AI is positioned to benefit from enterprise demand for autonomous workflow solutions. The timing aligns with broader market readiness for AI systems that can execute multi-step tasks without human intervention.
SPREAD AI develops engineering intelligence software that enables organizations to unify, analyze, and act on complex product data.
A production-hardened, domain-specific engineering ontology combined with in-place connectors and multi-agent AI apps that deliver immediate, domain-relevant value (requirements validation, error linking, product traceability) to OEMs without migrating data.
SPREAD centers product data in an explicit, permission-aware graph/ontology: entities (requirements, parts, functions, tests) are nodes with links, customer-specific subgraphs extend a shared core, and the platform exposes RBAC-style permissions and sync metadata. This is implemented as an authoritative graph database/knowledge layer that all apps and agents read from.
Emerging pattern with potential to unlock new application categories.
The system retrieves domain artifacts (requirements, past projects, CAD/PLM records) from the connected ontology and source systems to augment downstream generation/analysis. Incoming natural language requirements are grounded against retrieved engineering records and traces — a classic RAG pattern using a structured, graph-backed knowledge source.
Accelerates enterprise AI adoption by providing audit trails and source attribution.
SPREAD exposes and ships multiple autonomous agents/apps (pre-built domain agents plus customer agents) that act on the shared ontology and tools. The platform coordinates these agents to perform tasks such as transforming requirements, tracing dependencies, and inspecting errors — an agentic, tool-enabled architecture.
Full workflow automation across legal, finance, and operations. Creates new category of "AI employees" that handle complex multi-step tasks.
The platform ingests free-form requirements and converts them into structured artifacts, rules, and links within the ontology. This is NL-to-structured-representation (NL-to-code/rules) — parsing specifications into machine-actionable entities, validations, and trace links.
Emerging pattern with potential to unlock new application categories.
Multi-agent apps and user-defined agents operate over a shared central ontology. Agents and apps read/write the canonical graph and connectors provide source-of-truth access to enterprise systems. The content provides no details on model-to-model handoffs or model ensembles.
Former Production Engineer at Mercedes; Senior Strategy Consultant at Porsche. Listed as leading Product & Sales for SPREAD.
Previously: Mercedes-Benz, Porsche
Founder's automotive industry background and experience with Mercedes/VW/BMW alignment with SPREAD's focus on automotive engineering data, PLM integration, and enterprise-scale platform adoption. Strong alignment between founder experience and product value proposition.
partnership led
Target: enterprise
custom
field sales
• Mercedes-Benz, VW, BMW as early customers
• European OEM references and industry press coverage
Unified engineering data platform enabling end-to-end product lifecycle traceability and rapid decision-making via a shared ontology across requirements, design, and manufacturing
Applying one engineered, versioned ontology across every engineering domain (requirements, software, hardware, tests, field) and using it as the primary substrate for agents and apps is relatively rare in industry — especially with claimed live sync to many enterprise systems.
Layered customer subgraphs allow rapid, governed customization without forking the core model — enabling both reuse across customers and strict control of the canonical schema.
SPREAD AI operates in a competitive landscape that includes Siemens (Teamcenter / Xcelerator), Dassault Systèmes (3DEXPERIENCE / ENOVIA), PTC (Windchill).
Differentiation: SPREAD positions as an ontology-first layer that reads/writes in place across PLM/CAD/ERP (including Teamcenter) without migration, offering domain-specific AI apps (Requirements Manager, Error Inspector) and multi-agent AI on top of a pre-built engineering ontology rather than replacing the PLM.
Differentiation: SPREAD focuses on connecting and normalizing data across multiple vendors' systems (including 3DEXPERIENCE) into one shared ontology and providing low-code apps and AI agents that operate on that unified graph, instead of being a single-vendor suite that expects customers to migrate to its stack.
Differentiation: SPREAD differentiates by offering in-place mapping (no replatforming) across Windchill and other systems, plus engineered, pre-built domain ontologies and AI-driven applications for earlier error detection and requirements validation tailored to regulated industries.
Graph-first engineering ontology that is explicitly designed to span PLM, CAD, ERP, ALM, MES, simulation and test systems — not just a metadata index but a live model where 'built-in stays, custom extends' via a customer subgraph layering model. That combination (core canonical graph + per-customer subgraph in the same layer) is a concrete product decision that changes upgrade/extension semantics vs. simple federation.
In-place, bidirectional connectors rather than full data migration. They claim read/write adapters into Teamcenter, Windchill, 3DEXPERIENCE, SAP, etc., leaving the source systems live. That's an unusual tradeoff: minimize migration costs and user friction but push enormous complexity into adapter logic (transaction semantics, conflict resolution, partial writes, versioning).
Multi-agent AI layer that operates directly on the shared ontology. Instead of separate agent-specific representations or ephemeral caches, agents, apps, and UIs all read/write the same model — enabling automated cross-domain workflows (e.g., a Requirements Manager agent that links a sentence to CAD parts and to simulation tests in one transaction).
Real-time and safety-domain signal integration implied by the showcased BMS requirements. They are mapping requirements (e.g., 10Hz sampling, 50ms disconnect for isolation faults, ASIL-C traceability) to CAN/FlexRay signals and linking them to simulation and test cases. That means their graph must hold time-series/telemetry references and truth about hard real-time constraints and regulatory artifacts.
Low-code app generation (SPREAD Studio) that builds interactive, traceable apps on top of the ontology, not separate dashboards. This indicates they treat the ontology as an application substrate rather than merely a data layer.
If SPREAD AI achieves its technical roadmap, it could become foundational infrastructure for the next generation of AI applications. Success here would accelerate the timeline for downstream companies to build reliable, production-grade AI products. Failure or pivot would signal continued fragmentation in the AI tooling landscape.
“Run AI at domain-expert level.”
“One ontology across requirements, simulation, software, hardware, production, and the field.”
“L03 Apps & agents Requirements Manager, Product Explorer, Error Inspector, plus your own agents.”
“All read from one ontology, so they always agree.”
“From automotive to aerospace — deploy AI-powered solutions across the full product lifecycle.”
“2024 Multi-agent AI introduced.”
