Landscape

The Convergent Pattern

When independent vendors arrive at the same workflow shape, that is the architecture announcing itself. The question is no longer whether AI belongs in MBSE — it is whether the tool was built around the pattern or had it bolted on after.

Greg DogumMay 202611 min read

On April 26, 2026, MathWorks Consulting Services published a blog post walking through an AI-native MBSE workflow built on Claude Code, the MATLAB MCP server, and System Composer. The post is generous and well-written. It describes a five-step interaction loop the agent uses for every architectural change: propose, approve, generate, run, confirm. The AI drafts a plan in plain language. The engineer reviews and requests changes. The AI then writes a script that creates the actual model artifact. The script executes against System Composer. The output is shown back to the engineer for confirmation before anything else happens.

That five-step loop is, structurally, the same pattern Cairn was being built around at the same time — under a different name, in a different stack, against a different data model. We call it the ChangeSet review pattern: the AI proposes a structured set of operations, a validator checks them, a human accepts or rejects each one, and only the accepted operations are applied. Different vocabulary. Same shape. When two teams arrive at the same workflow shape from opposite directions, that is not a coincidence. That is the architecture announcing itself.

The right pattern for AI inside a high-consequence engineering tool is being independently rediscovered by everyone who builds in the space. The interesting question is no longer what the pattern is. It is whether your tool was built around it — or had it bolted on after.

What the Last Six Months Have Made Visible

In the second half of 2025 and the first half of 2026, three independent moves established that AI-MBSE is no longer speculative. Every one of these signals is real, well-resourced, and pointing in the right direction. Read in sequence, they tell a story about where the category is going and why the architecture matters.

Three Signals · 2025–2026
Real moves, real money, all converging on AI inside the engineering loop
Nov 2025
CoLab — $72M Series C
EngineeringOS positioning. AutoReview AI peer checker. 47,000-engineer waitlist. Pitch: AI agents on top of CAD design review collaboration.
CAD-adjacent · enterprise
2025–2026
ThunderGraph — Automating MBSE
AI graph agent that builds SysML v2 model elements incrementally, traversing the graph by BFS to verify against source documents and standards.
SysML v2-native · enterprise
Apr 2026
MathWorks — Agentic AI for System Composer
Claude Code + MCP + System Composer. RFLP workflow. Five-step propose/approve/generate/run/confirm loop. Public skills repository.
incumbent stack · agentic
Q1 2026
SysON, Innoslate, others
SysML v2-native authoring with embedded AI assistants. Some open-source. Most still in beta. All converging on similar workflow shapes.
v2-aligned · maturing

Read those four together. The vendors are different. The stacks are different. The customer profiles are different — CoLab sells into hardware companies that already have CAD design review processes, MathWorks sells into engineering organizations already on Simulink and System Composer, ThunderGraph sells into defense and aerospace programs adopting v2. None of them are direct competitors. But every one of them is building toward the same architectural pattern: an AI agent that proposes structured changes to engineering artifacts, with a human-in-the-loop review gate before anything becomes part of the model.

When Convergence Is the Validation

The cleanest way to see what is happening is to put two of these workflows side by side. On the left, the loop MathWorks describes for their Claude Code + System Composer integration. On the right, the loop Cairn has been running since launch. They were designed independently, by different people, against different data models. The shape is identical because the shape is what works.

The Convergent Loop
Independent designs, identical shape
MathWorks · Apr 2026
"Agentic AI for MBSE" workflow
1
Propose
AI drafts the change in plain language
2
Approve
User reviews, requests changes
3
Generate
AI writes the script to enact the change
4
Run
Script executes against the model
5
Confirm
Output shown back, user confirms
Cairn · since launch
ChangeSet review pattern
1
Propose
Specialist drafts a structured ChangeSet
2
Validate
Schema and consistency checks
3
Review
User accepts/rejects each operation
4
Apply
Accepted ops mutate the model
5
Audit
Every applied op is replayable, undoable
Same five steps. Different vendors. Different stacks. The pattern is being independently rediscovered because it is the right pattern. AI proposes, the human reviews, the model changes only after approval, every change is auditable. That is what governed AI inside an engineering tool actually requires.

ThunderGraph's BFS-based incremental graph construction is a slightly different shape — they build a single element at a time and verify by graph traversal — but it lands in the same place. CoLab's AutoReview annotates and surfaces, leaving the human as the authority on what gets changed. SysON's emerging AI features follow the same "AI proposes, human disposes" pattern. Every serious entrant in this category has converged on a propose/review/apply governance loop, because anything else creates an AI that silently mutates engineering artifacts — and no engineering organization will accept that.

Bolt-On vs Built-In

If everyone is converging on the same loop, why does the architecture matter? Because the loop is the visible part. What sits underneath it — the data model, the validator, the audit trail, the way undo and redo work, the way partial acceptance is handled — is where bolt-on and built-in start to feel different.

A bolt-on architecture starts with an existing tool that was designed for a single human user — System Composer, Cameo, an existing CAD review platform — and grafts the AI workflow onto it. The propose step works because the AI can be prompted. The review step works because the human can be shown a draft. The apply step works because the underlying tool already has APIs for changing the model. But the seams underneath show up the moment you stretch the workflow — when a ChangeSet has fifteen operations and the user wants to accept ten, when a partial acceptance leaves the model in a state the original tool was never designed to validate, when the audit log needs to capture not just what changed but who proposed what and why, when the undo needs to roll back across multiple specialists.

A built-in architecture starts from the assumption that AI is a first-class author. The data model is shaped by it. ChangeSets are not a UI overlay — they are the canonical representation of every change, regardless of whether a human or an AI proposed it. The validator runs before any ChangeSet reaches the user, not as a wrapper around the apply call. Partial acceptance is a property of the data model, not a feature retrofitted into the modal. Audit is automatic because it is the same shape as the applied change. When AI is a substrate decision rather than a feature decision, the seams that bolt-on architectures keep stumbling over simply do not exist.

The question
Bolt-on answer
Built-in answer
01
User accepts 10 of 15 ops in a ChangeSet. What happens to the other 5?
Bolt-on
Either re-prompt the AI to regenerate, or apply all-or-nothing. Partial acceptance is a UX problem the underlying tool wasn't designed for.
Built-in
Apply the 10 accepted ops. Discard the rest. The data model treats each op as independent and re-validates the result. No regeneration required.
02
Six months later, why was this requirement added?
Bolt-on
A separate audit log, often partial. The connection between AI rationale and the resulting model element relies on bookkeeping outside the tool's native data structures.
Built-in
The ChangeSet that introduced it is the audit trail. The AI's rationale, the user's review decision, and the resulting op are the same artifact. Replay is intrinsic.
03
The user edits a node manually. Does AI know?
Bolt-on
Often no. The manual path and the AI path live in different abstractions. Subsequent AI operations may overwrite hand edits or drift from them.
Built-in
Manual edits go through the same ChangeSet pipeline, with the user as proposer. AI sees them as part of the model history. Drift is structurally avoided.
04
What does undo do?
Bolt-on
Whatever the underlying tool's undo stack supports. Often partial across AI and manual edits. Sometimes loses rationale.
Built-in
Inverse of the last applied ChangeSet. Carries rationale forward. Symmetric with redo. Same shape regardless of who proposed.

None of this is a knock on the bolt-on architectures. They are doing real work and they are shipping real value. MathWorks, in particular, is doing something subtle and important — they are showing their existing System Composer customers what an agentic workflow looks like, in a stack those customers already trust. That is exactly how you bring a conservative engineering audience along. The hedge in the post — "nobody is going to design an entire system with agentic AI in one session" — is honest, and it reflects the constraints of the bolt-on path. When AI is a layer on top of a tool designed for a different era, the demos are real but the depth has natural limits.

Why Day-Zero Architecture Wins the Long Game

The AI-MBSE category will not be won by whoever has the best agent prompt. The agents are converging on the same shape, and the foundation models underneath are commoditizing every quarter. What will differentiate winners from also-rans is the foundation underneath the agent — the parts that determine whether the workflow holds together when a real engineer uses it on a real program for a real year.

That foundation is harder to retrofit than to design. The data model that treats AI proposals as first-class citizens. The validator that runs before, not after. The audit that is structural, not bolted on. The partial acceptance, the cross-specialist coordination, the unified history of human and AI changes. Every one of these is solvable in a bolt-on architecture, but solving them later costs more than designing for them earlier — and the engineers who feel the seams first will be exactly the practitioners the category most needs to win over.

01
The Validation
Convergence is the architecture announcing itself
When MathWorks, ThunderGraph, CoLab, and Cairn arrive at the same propose/review/apply loop from four different starting points, that is not coincidence. It is the structure of the problem forcing everyone to the same answer. The question stops being "what should the workflow look like" and becomes "how deep does the workflow go in your system."
02
The Difference
Bolt-on shows up in the seams
Partial acceptance, audit, manual-edit reconciliation, cross-specialist coordination — these are the questions where bolt-on and built-in feel different. Demos rarely expose them. Real use does. The architectures that handle these natively will produce the workflows engineers actually trust over a multi-year program.
03
The Position
Built around the pattern, not built to host it
Cairn was not retrofitted with AI. It was designed around the assumption that AI would be a first-class author and governance would be the substrate. The convergent pattern emerging across the category validates the bet. The seams that bolt-on architectures will keep working through are not seams in a system built around them from day zero.

The good news for the category is that the right pattern has been found. The good news for engineers is that they will get to choose between several real implementations of it over the next few years. The architecture question — bolt-on or built-in — will be the one that decides which of those implementations age well, which start to feel like the right tool for the next decade, and which feel like the right tool only for the demo.

Cairn is making one bet on that question. The convergent pattern says the bet was a good one to make.

Cairn is the AI engineering workbench for systems that matter.

Sign up and start building for free.