AI Summary
5 min readWhen AI Writes Code but Can't Run the System
Jad Naous, CEO of Grepr and a systems engineer who built distributed systems at AppDynamics and Imply before founding his own company, has a clear-eyed view of where AI excels and where it stalls. His 10-person team uses coding agents to ship production code at speeds that would have taken months just a few years ago. But when the conversation shifts from writing software to running it—operating distributed clusters, troubleshooting incidents, optimizing live systems—the picture changes dramatically. The episode is a grounded debate between Naous and a16z GP Martin Casado about whether AI's success in code generation will extend to operations, or whether the two problems are fundamentally different in kind.
The 10x Engineer Just Got 100x
Naous describes a workflow that would have been unthinkable two years ago. He uses Claude Code with Opus 5, running on Windows Subsystem for Linux, and has learned to manage multiple agent sessions simultaneously—typically three or four—that interact through the command line rather than through subagents that burn tokens on research. His trick: tell Claude to spin up command-line subagents that maintain persistent sessions, so they can review each other's PRs and iterate on design documents together.
Continue reading the full summary in the app — free to try.
Read Full Summary →Free • No credit card required
Never miss an episode of a16z Show
Get every new episode summarized in your inbox — free, ~5 minutes to read.
No spam. Unsubscribe anytime.
What you'll learn
- 1 (00:00) **The Core Divide: Digital vs. Physical** - The episode opens by framing its central tension: AI thrives in the digital world, but the "actual world" of running systems is fundamentally different.
- 2 (02:52) **Jad's Background: From Systems Architect to CEO** - Jad provides a brief history of his career, establishing his credibility as a deep systems thinker.
- 3 (07:31) **The Origin of Grepr: Solving the Observability Cost Crisis** - Jad explains the market problem that led to his company, focusing on the broken economics of observability.
- 4 (11:16) **How AI Changes Engineering: The Leverage Gap** - Jad shares his direct experience using AI on his small team, highlighting a critical divergence in engineer performance.
- 5 (18:30) **Jad's Workflow: Claude Code, Sub-Agents, and 10x Gains** - Jad details his personal, highly optimized process for using AI to ship production code.
- 6 (24:19) **The "10-Day Project" That Used to Take 6 Months** - Jad provides a concrete example of the massive productivity gains from his AI workflow.
- 7 (26:54) **The Unanswered Question: Can AI Reason About Operations?** - The conversation pivots from writing code to the much harder problem of running it in production.
+ Full timestamped outline available in the app
Guests on this episode
Show Notes
a16z General Partner Martin Casado sits down with Grepr CEO Jad Naous to ask what comes after AI learns to write code: can it also debug and operate the systems that code creates?
Jad shares how AI has transformed engineering inside his 10-person team, cutting some projects from six months to just 10 days. But operating production systems is a harder challenge, with unpredictable workloads, failures, and countless possible states.
They discuss whether AI can learn to diagnose problems, test hypotheses, and debug systems like a human engineer. They also explore AI SREs, security, query optimization, and why the best engineers may be gaining the most leverage from coding agents.
Resources:
Follow Jad Naous on X: https://x.com/jadtnaous
Follow Martin Casado on X: https://x.com/martin_casado
Learn more about Grepr: https://www.grepr.ai/
Stay Updated:
Find a16z on YouTube: YouTube
Find a16z on X
Find a16z on LinkedIn
Listen to the a16z Show on Spotify
Listen to the a16z Show on Apple Podcasts
Follow our host: https://twitter.com/eriktorenberg
Please note that the content here is for informational purposes only; should NOT be taken as legal, business, tax, or investment advice or be used to evaluate any investment or security; and is not directed at any investors or potential investors in any a16z fund. a16z and its affiliates may maintain investments in the companies discussed. For more details please see a16z.com/disclosures.
Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.