Founded
2024
Aircraft are still assembled by hand, one hole at a time
Airframe structures are the most precisely engineered objects in serial production, and they are joined by people with feeler gauges, shim stacks and clipboards. Every hour of that is an hour the industry cannot afford at the rate it has committed to build.
Team
68
▲ +89% increase
Stations live
70
▲ +46% increase
Holes inspected
4.2M
▲ +61% increase
Every fastener, flawless fit
Rivetira exists to make aircraft structural assembly autonomous, conforming and traceable — so that build rate is limited by demand rather than by how many people can hand-fit a wing box.
That means three things in practice. Every characteristic is measured rather than sampled. Every join is predicted before the parts meet. And every operation writes a record that survives the program.
The assembly line that learns
A factory where the thousandth airframe is measurably better than the first, because the line remembers every hole it has ever drilled.
Craft knowledge in aerospace assembly currently lives in the hands of mechanics who are retiring faster than they can be replaced. Encoding that knowledge is not a cost-reduction exercise. It is how the industry keeps the capability at all.
Why assembly, and why now
Three curves crossed at roughly the same time. Rivetira exists in the space between them.
| Year | Committed build rate | Skilled assembly workforce | Cost of edge inference |
|---|---|---|---|
| 2019 | 100 | 100 | 100 |
| 2020 | 62 | 91 | 74 |
| 2021 | 71 | 88 | 51 |
| 2022 | 88 | 89 | 34 |
| 2023 | 104 | 91 | 21 |
| 2024 | 121 | 90 | 13 |
| 2025 | 138 | 88 | 8 |
| 2026 | 154 | 86 | 5 |
How we decide things
Measure before you claim
Every result we publish has a baseline that was signed before we touched the station. A number without a baseline is a story.
Autonomy is earned, never granted
No agent moves up a level because a model is confident. It moves up because measured accuracy cleared a gate and a human signed.
Software never sits on the safety path
The stop is hardware. If our runtime disappeared mid-operation, nothing unsafe would happen. That constraint shapes the architecture.
Model the structure, not the process
Processes vary by plant and by mechanic. The physical structure does not. Modelling geometry generalises; modelling procedure does not.
Build for the mechanic, not the executive
If the person at the station does not trust the recommendation, autonomy stalls at advisory forever. The station UI is the real product surface.
Publish the failures
Rollback rates, weak material stacks and gate failures are in our benchmarks. Hiding them would make the wins meaningless.
The people building it
Ilse Vandermeer
Co-founder & CEO
Fifteen years in final assembly operations, including two narrowbody rate ramps. Left a plant director role because the fifth rework meeting of the week was about the same join.
Tobias Ferrand
Co-founder & CTO
Robotics and perception. Built drilling control systems that shipped on three aerostructures programs before deciding the control loop was the wrong place to stop.
Dr. Amara Osei
Head of Research
Applied mechanics and tolerance modelling. Her thesis on gap-field prediction from as-built geometry is the direct ancestor of the predictive shimming agent.
Rune Halvorsen
VP Engineering
Edge infrastructure at industrial scale. Spent a decade making software behave predictably in buildings with cranes, forklifts and no reliable network.
Priya Chandrasekar
Head of Quality & Airworthiness
Former chief quality engineer on a defense program. Joined to build the audit trail she spent years assembling by hand.
Marcus Oyelaran
VP Field Operations
Manufacturing engineering across four OEMs. Runs the line assessment methodology and refuses to let a deployment start without a signed baseline.
[PLACEHOLDER: team members are representative of Rivetira's founding and leadership profile.]
What the team is made of
| Segment | Share |
|---|---|
| Robotics & controls | 16 |
| Perception & ML | 14 |
| Platform & edge | 11 |
| Field & deployment | 12 |
| Quality & airworthiness | 5 |
| Product & design | 4 |
| Go-to-market & ops | 6 |
| Quarter | Headcount |
|---|---|
| 2024 Q3 | 6 |
| 2024 Q4 | 11 |
| 2025 Q2 | 24 |
| 2025 Q4 | 41 |
| 2026 Q2 | 58 |
| 2026 Q3 | 68 |
What is actually hard to copy
- STEP 01
Physical process data
Four point two million inspected holes with measured geometry and known outcomes. This data does not exist anywhere else and cannot be simulated into existence.
Inspected characteristics4.2M
- STEP 02
Certified process envelopes
Qualified process windows per material stack, per structure, validated against real conformance outcomes. Each one took a program to earn.
Qualified envelopes112
- STEP 03
Deep OT integration
Twenty-four machine classes driven directly, nine metrology SDKs integrated. Each integration is weeks of unglamorous work no model can shortcut.
Machine classes24
- STEP 04
Earned trust
Forty-seven stations running closed loop on flight structure. The reference that gets the forty-eighth is not something a competitor can buy.
Closed-loop stations47
Assembly autonomy, measured on the line
Every figure below is produced by the same telemetry the agents act on — station cycle, hole quality, gap field, fastener state and conformance. Pilot and design-partner aggregate, trailing 12 months.
Holes drilled under agent control
41.6M
▲ +32% increase
Right-first-time, structural joins
97.9%
▲ +9.1% increase
Shim hours removed per join
68%
▼ -68% decrease
Assembly-line uptime
99.94%
▲ +0.3% increase
Aggregate across design-partner lines. Baselines are the same stations before Rivetira, measured over an equivalent period.
What changes on the floor
Plant directors, liaison engineers and airworthiness leads on what autonomy did to their numbers.
“We stopped arguing about whether the gap was 0.4 or 0.6 millimetres. The twin predicted it, the shim came off the machine right, and the join closed in one pass.”
−64% shim hours per join
“The rate ramp was going to cost us four more positions. Instead the line rebalanced itself every shift and we found the capacity inside the stations we already had.”
+38% delivered rate
“Every hole is now inspected, not one in four. My airworthiness record writes itself, and I can hand an auditor a complete trace in ninety seconds.”
100% inspection coverage
Airframes assembled on Rivetira across commercial, defense, business-jet, eVTOL and space-structure programs: 8 factories, 70 assembly stations.
Meridian Aerostructures
Calder Aviation Group
Northvane Defense
Solstice Jets
Helion Air Mobility
Orbital Frame Systems
Built for programs that cannot leak and cannot fail
Airframe geometry is among the most protected IP in manufacturing, and much of it is export-controlled. Rivetira is architected for that reality from the edge up.
Certified
SOC 2 Type II
Annual audit covering security, availability and confidentiality of the control plane.
Supported
ITAR / EAR aware
US-person access controls, on-prem and air-gapped deployment for controlled programs.
Mapped
AS9100 alignment
Quality records, NCR flow and traceability mapped to aerospace quality requirements.
In progress
ISO 27001
Information security management system certification underway. [PLACEHOLDER: target date]
Data stays where the airframe is
- Factory edge runtime holds all program geometry, imagery and metrology locally
- Air-gapped mode: model updates arrive by signed media, telemetry never leaves
- Per-tenant isolation and per-program cryptographic separation
- Encryption in transit and at rest; customer-managed keys available
Assurance-grade by default
- Immutable audit trail for every agent action, sensor reading and human override
- SSO/SAML, SCIM provisioning and role-based access down to station level
- Fail-safe stop for drills, fasteners, crawlers and cranes at < 40 ms
- Human-in-the-loop mandatory for airworthiness-critical dispositions
Questions engineers actually ask
Anything else goes to the people who build it. Ask an assembly engineer or read the full FAQ.
No. We build the intelligence layer that runs on robots other people build. Owning the hardware would slow us down and would make us compete with the machine builders who are our best distribution.
[PLACEHOLDER: funding details available on request.] We are deliberate about capital because deployment pace is limited by accreditation and trust, not by cash.
Because you cannot build this from a distance. The founding team wanted to be close enough to a real final assembly line to walk onto it, and most of the engineering team has spent time at a station this quarter.
Come build the assembly line that learns
A line assessment maps one station, quantifies the rework, shim and rate opportunity, and returns a modelled ROI in three weeks. No production disruption.