On episode 9 of Third Loop, the Progressive Delivery team asks whether AI can genuinely support creativity without flattening everything into the same familiar output. Together, they unpack what it would mean to treat AI as one useful tool rather than the answer to every problem.
VC partner from Code ➡️ Company. Hands-on from Day 0 with the world’s best technical founders building the future of enterprise infrastructure.
- Agents are good at tasks. They're still bad at owning outcomes, and the gap isn't a model problem. New from @amirzo: the five systems we're missing before agents can go from order takers to outcome owners. heavybit.com/library/articl…
- In episode 54 of The Kubelist Podcast, David Crawshaw (@davidcrawshaw) shares how a weekend WireGuard experiment became @Tailscale, and how a series of unsuccessful developer-tool experiments eventually became @ssh_exe_dev. The conversation offers a candid look at product
- On episode 3 of Lab Notes, Amir Zohrenejad (@amirzo) speaks with Qizheng Zhang (@qizhengz_alex) about the future of AI beyond model scaling. From context engineering and Meta-Harness to autonomous research agents and benchmark design, they explore how recursive self-improvement
- On episode 41 of @opensourceready, @bdougieYO and @johncodes sit down with Phil Estes (@estesp). They explore why the definition of a container remains surprisingly fuzzy, how microVM-backed sandboxes could support the next generation of AI agents, and what AI-generated

