Two engineers rebuilt a FedEx supplier platform in 3.5 months with repository graphs
alex_verem · x · 2026-07-24
- Two forward-deployed engineers rebuilt a live FedEx supplier’s delivery orchestration platform in 3.5 months, versus an original estimate of 7–8 months.
- Their main advantage was not faster code generation, but a repository-wide knowledge graph that mapped modules, dependencies, data flows, and domain language.
- The team documented the entire codebase in markdown first; agents read that map before writing code, so generated changes matched the real architecture and fit inside context windows.
- Humans still reviewed all 122 merged pull requests, while the models wrote about 90% of the code.
- The post argues that with the right infrastructure, two engineers can now produce output that used to require a much larger team.
More from coding & agent
- Model Routers Can Break Context Cache in Multi-Turn Chats, Incurring Penalties — pvncher · 2026-07-24
- AI Agent Demonstrates Autonomy by Generating Long-Term Goals Unprompted — QuixiAI · 2026-07-24
- Developer intelligence dashboard spots one engineer spending $20,145 on AI in 30 days — alex_verem · 2026-07-24
- Offloop says its 3B dispatcher beats Codex and Claude Code on GDPval at $1.65 a task — dr_cintas · 2026-07-24
- Codex on a Cloud VM: Unleash the AI Coding Agent — paw_lean · 2026-07-24
- Codex Hits 5M Weekly Users; AI Model Eval & Transparency Discussed — altryne · 2026-07-24