Skip to content

Latest commit

 

History

History

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
 
 

README.md

Driverless — Case Studies Index

This directory contains case study documents for each module. Every case study follows the same structure:

  1. What Happened at Uber — the specific problem, incident, or design failure
  2. Impact — user trust, revenue, legal consequences
  3. Root Cause — technical or organizational
  4. Our Approach — how Driverless solves it differently
  5. Why It Works — references to research, industry best practices
  6. Trade-offs — what we sacrifice and why it's acceptable

Phase 1 — Core Platform & Identity

Module Case Study Status
User Management Monolith scaling pain & the 2016 data breach ✅ Complete
Authentication Account takeover attacks ⬜ Planned
Driver Onboarding Background check failures & impersonation ⬜ Planned
Profile & Privacy GDPR violations (€290M + €10M fines) ⬜ Planned

Phase 2 — Real-Time & Geospatial

Module Case Study Status
GPS Tracking Write amplification at scale ⬜ Planned
Geospatial Indexing O(n) nearest-driver to O(1) cell lookup ⬜ Planned
ETA Calculation Straight-line distance inaccuracy ⬜ Planned
Routing Unsafe route suggestions ⬜ Planned

Phase 3 — Matching & Dispatch

Module Case Study Status
Ride Lifecycle Double-dispatch race conditions ⬜ Planned
Matching Algorithm "Closest driver" failure modes ⬜ Planned
Batched Matching Greedy vs global optimization ⬜ Planned
Supply Prediction Reactive surge vs proactive repositioning ⬜ Planned

Phase 4 — Pricing & Payments

Module Case Study Status
Dynamic Pricing Surge pricing opacity & Hurricane Sandy ⬜ Planned
Fare Estimation Unpredictable final fares ⬜ Planned
Payments Cash markets & payment fraud ⬜ Planned
Driver Earnings Opaque commission & driver churn ⬜ Planned
Promotions Mastermind rules engine & abuse rings ⬜ Planned

Phase 5 — Safety & Fraud

Module Case Study Status
In-Ride Safety Reactive vs proactive safety engineering ⬜ Planned
Identity Verification Driver impersonation incidents ⬜ Planned
Fraud Detection (RADAR) Adversarial fraud & ML-powered detection ⬜ Planned
Ratings Rating manipulation & accountability ⬜ Planned
Disputes Manual resolution doesn't scale ⬜ Planned

Phase 6 — Scale & Compliance

Module Case Study Status
Notifications Multi-channel delivery reliability ⬜ Planned
Compliance Engine Legal battles across jurisdictions ⬜ Planned
Observability Silent outages & "design for failure" ⬜ Planned