a16z: AI Can Write Code but Still Can't Operate Production Systems
a16z · youtube · 2026-10-16
In an a16z podcast, Grepr CEO Jad Naous and GP Martin Casado discuss the limits of coding agents in real engineering.
- Writing code is easy, operating it is hard: AI can produce production code in minutes, but real systems have unpredictable workloads and countless failure states, making debugging and operations a fundamentally harder challenge.
- Measured gains: Inside Jad's 10-person team, coding agents have reshaped engineering, turning projects that once took six months into roughly 10 days.
- Reasoning about ops: They explore whether AI can troubleshoot production systems by detecting changes, forming hypotheses, and running experiments.
- Takeaway: The best engineers may gain the most leverage from AI.
- Also covers why observability got so expensive, how Grepr catches incidents early, and why security is the harder problem.
Related event: a16z Podcast: AI Excels at Coding but Struggles with Production Ops(2 posts)→
More from coding & agent
- StackOne ships Tool Policies: fine-grained guardrails for MCP tool calls — afp-media · 2026-10-17
- Huntley: developers displaced by AI 'failed a curiosity test' — rseroter · 2026-10-17
- Teaching may be the best prerequisite for prompting agents, argues Geoffrey Litt — sh_reya · 2026-10-17
- RAX open-sources everything: zero demos, all experiment sessions included — du_yilun · 2026-10-17
- RAX separates the coding agent workspace from the robot stack for multi-embodiment — du_yilun · 2026-10-17
- shadcn reflects: it's only been a year since Claude Skills launched — shadcn · 2026-10-17