Team Ladebuggers
PlatInform
Kearney PERLab x UIUC AI Entrepreneurship Hackathon
Platforming has two problems, and they feed each other.
It takes real engineering effort.
Every product decision has to account for what the company already designed and bought, across every team.
43percent
name engineering effort as the top challenge
Kearney PERLab survey of 100 leaders, November 2026. docs/CONTEXT.md line 9.
Most companies do not see the potential.
Companies running a platforming strategy at scale, by industry
- Aerospace and defense80
- Consumer electronics75
- Auto and transport71
- Construction57
- Industrial equipment44
- Medical43
- Telecom38
- Energy and power20
Our target market
Kearney Platforming Survey, 2026.
Both problems are the opportunity: more companies to bring into platforming, and engineering work that continues after the first project.
Platforming means making every product use the same parts.
Platforming means never losing a design the company already proved.
One product · VRK-1000
- Half Bridge IGBT Module 600V 50A
- Half Bridge IGBT Module 1200V 100A
- Isolated Gate Driver Board 6 Channel
- Electrolytic DC Link Capacitor 470uF 450V
- Film Capacitor 20uF 800V
- Bleeder Resistor 10k 10W
VRK-3000
VMX-20
VSI-5
VSI-10
VBS-50
5 more products stack onto it
Power stage, proven in 2019
Half Bridge IGBT Module 600V 50A
Half Bridge IGBT Module 1200V 100A
Isolated Gate Driver Board 6 Channel
Electrolytic DC Link Capacitor 470uF 450V
Film Capacitor 20uF 800V
Bleeder Resistor 10k 10W
Gate driver board and its isolation barrier
Busbar drill pattern and creepage layout
Dead time settings and the short circuit test
Capacitor clamp and terminal pitch
Bleeder network and the discharge time proof
Pre-charge sequence in firmware
Reuse the block, inherit the brackets, harness, firmware, and tests. Start from there.
Design concept: a shared power stage module, the same block built into 6 products.
Sample data. data/derived/voltara_interfaces.json, the two power stage records.
PlatInform sits where product decisions get made.
- It works inside the web tools teams already use for product development, shown here on Velocity AI Studio.
- It answers from the company's own purchasing and design data.
- It points to proven designs to build on, and to single parts as well.
How we know part A can replace part B.
The question a judge, BK, asked us on Saturday.
10-92550
SL5-2214B
Fellegi-Sunter record linkage (1969): weighted fields, two thresholds, human review between them.
0.00Fan vs wire fan guard
0.68FAN - SEE DWG
0.88Two-input vs three-input board
1.00The same fan
of matches correct in a blind test on 252 part numbers
1 mistake in 602.
When two products share no parts
VRK-3000 and VSI-10 share 0 of their 89 part numbers, and both are built on the same half bridge power block.
Part matching finds identical parts. Design matching finds what different products have in common.
Next
Planned
Reading datasheets and drawings for parts that text cannot describe.
Anthropic, Testing Claude on real-world electronics design, November 2026.
PlatInform sits where product decisions get made.
- It works inside the web tools teams already use for product development, shown here on Velocity AI Studio.
- It answers from the company's own purchasing and design data.
- It points to proven designs to build on, and to single parts as well.
How we know part A can replace part B.
The question a judge, BK, asked us on Saturday.
10-92550
SL5-2214B
Fellegi-Sunter record linkage (1969): weighted fields, two thresholds, human review between them.
0.00Fan vs wire fan guard
0.68FAN - SEE DWG
0.88Two-input vs three-input board
1.00The same fan
of matches correct in a blind test on 252 part numbers
1 mistake in 602.
When two products share no parts
VRK-3000 and VSI-10 share 0 of their 89 part numbers, and both are built on the same half bridge power block.
Part matching finds identical parts. Design matching finds what different products have in common.
Next
Planned
Reading datasheets and drawings for parts that text cannot describe.
Anthropic, Testing Claude on real-world electronics design, November 2026.
All of this runs on a company's own data.
Today no team has a reason to platform, so no company hands that data over.
PlatInform gives every function a reason, in three steps.
Eight questions about the company
1 of 3
Step one: a two-minute ballpark for the C-suite.
Eight plain questions give executives a rough Platforming Index and a rough cost. For Kearney it is a business development opening.
Questions shaped with Andrew Cecola of Kearney PERLab.
The measured index, read from the company's own parts list
2 of 3
Step two: the measured index.
With the bill of materials in, the company sees its real score, where the duplicates are, and what they cost, explained by an assistant for each role.
- CEO28 of 100index points
- CFO$1.13Mdollars a year
- Engineering13,700engineering hours a year
- Supply chain12 and 192suppliers and part numbers
- Sales3 to 14weeks of lead time
Sample company: Voltara, a fictional energy and power manufacturer, the sector at 20 percent.
Decisions recorded where the functions disagree
3 of 3
Step three: every function owns a piece.
Leaders see where they disagree and decide who owns what. Each function gets a reason to act in its own unit.
Each decision goes back into the data the next answer is read from.
Every decision improves the data and the plugin.
Two leadership decisions: index 28 to 66, no part changed.
How Velocity AI Studio shaped PlatInform.
It validated the problem: 21 sources and 29 research chats, every figure cited.
It changed our claim: we planned to say companies with duplicate parts fail to platform. Velocity found no data for that, so we dropped it.
It is where the plugin runs: our browser extension was built on Velocity AI Studio.
Why this is a business.
It takes real engineering effort.
Sustained work.
Platforming is never one project. Each product line, each acquisition, and each new design brings more to measure and more to reuse. The plugin keeps the client engaged between projects.
$1.13Ma year
found on $39.33M of parts spend in a 6-product sample, about 3 percent.
Most companies do not see the potential.
Business development.
A two-minute ballpark gives any manufacturer a number to react to. Every result is a reason to start a conversation with Kearney.
Where the conversations start
Ballpark: free, opens the door.
Measured index: a paid diagnostic.
Plugin: a yearly license that keeps the gain.
PlatInform turns Kearney's own research into the first meeting, the first project, and the reason to stay.
Team Ladebuggers