a16z podcast: AI cut a 6-month feature to 10 days and widened engineer gaps 10x
a16z · x · 2026-10-16
a16z's Martin Casado talks with Grepr founder/CEO Jad Naous on whether AI can run production systems as well as it writes code.
Key points:
- Leaders spend 5-20% of infra budgets on observability with diminishing reliability returns
- Grepr sits in log streams, scans up to 1PB of logs/day, cuts observability costs 90%, and builds agents that use those signals to troubleshoot production issues
- At his 10-person company: a data lake feature once estimated at 6 months took 10 days with AI; the gap between okay and great engineers "just multiplied 10X"
- He runs 3-4 agent sessions in parallel, with a "tech lead" agent reviewing every PR
- Also covers why running live systems remains harder for AI than writing code
Related event: a16z Podcast: AI Excels at Coding but Struggles with Production Ops(2 posts)→
More from coding & agent
- steipete averages 423 merged PRs/day over 30 days with AI agent workflow — steipete · 2026-10-16
- Intent Mobile launches as first large-scale agent orchestration app optimized for foldables — LukeW · 2026-10-16
- Scoble demos Braid+Folio: local AI context sync across Macs, one-click site publishing — Scobleizer · 2026-10-16
- Pi Durable on Cloudflare: building hibernation-ready multiplayer chat agents on Durable Objects — ritakozlov · 2026-10-16
- vLLM ships Day-0 RL support for Beam in SkyRL, +62% LoRA decode throughput — vllm_project · 2026-10-16
- Opus 5.5 writes a 61KB single-file ocean sim in ~30 minutes, hitting 90fps at 1440p — Any-Reputation8118 · 2026-10-16