From Auto Parts Manufacturer to Productized Cloud Service at a Large Automotive Supplier
summary
A sub-10 person team inside a large auto-parts manufacturer built a smartphone-based safety product called yuriCargo that runs end-to-end from sales to operations and auto-detects driving events (hard braking/accel). Core detection algorithms failed in real-world trials, prompting a multi-year redesign that only stabilized after about five years; the project ran parallel customer pipelines using a sprinted “swimlane backlog” (70–80% focus per priority) and relied on an external BI tool that became a three-year technical debt. Recent work shows faster prototyping and validation of fixes using ChatGPT.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 3:36Product journey overview
- 4:20Customer support vs development
- 8:52Technology debt from rushed choices
- 11:41Core product flaws discovered
- 14:06Unstable customer interest
- 17:39Lessons and future improvements
- 19:05Conclusion and Q&A
related talks
16mShip Production Software in Minutes, Not Months
30mProduction › Prototype: How a Small Team Replaced a Core Platform in 90 Days
39mTransforming Car Development
46mNew Product Development Driven by Choosing Bad Practices and Failing Forward
18mAgile leadership and technical mastery, a harmonious alliance
25mContinuous compliance with Rust in automotive software
28mLearnings from building a CRM app at Google
30mShipping Through Chaos: Engineering Resilience Across Borders