--- schema: 1 type: video title: >- Prototyping as Leadership: How a CTO Ships with AI Agents — Hursh Agrawal, The Browser Company description: >- Fifteen recurring meetings a week, seven direct reports, a toddler at home, and somewhere inside that Hursh Agrawal ships between two and ten pull requests. It was not possible a year ago. His argument is that the manager schedule has quietly become usable as building time, which makes building part of a leadership job rather than a hobby squeezed around the edges of one. The reasoning is less about output than calibration. Frontier models turn over every few months and nobody can reliably tell you what a new one is good for, so a leader without hands on time ends up relaying opinions from Twitter. Building also solves a communication problem, since showing engineers a working prototype beats arguing about capability every quarter. He thinks leaders are unusually well suited to it as well, because they hold the most business context and their steering is therefore worth more per token than an engineer's. The mechanics come down to a single overnight loop. A coworker agent wired into Slack and the issue tracker spends twenty minutes assembling context into a prompt, that gets handed to a coding agent at five in the evening, and the morning block goes on reviewing whatever came back. He runs it to build features, to hill climb prompts against eval sets assembled from a handful of feedback dumps, and to train small classifiers end to end. The caveats are real. It only works on top of trustworthy CI, feature flags, and a prototype branch, and he is candid that his own code has caused incidents. Speaker info: - https://twitter.com/hursh - https://www.linkedin.com/in/hurshagrawal - https://www.hurshagrawal.com Timestamps: 0:00 - A manager's calendar, and why building became the job 2:44 - Judging a new model, and showing rather than arguing 4:25 - Execution is solved, judgment is not 5:16 - Four things leaders should build, and one to avoid 6:05 - The daily shape, and why 5pm matters most 6:56 - Building a feature overnight, and assembling the prompt 9:28 - Verification, PR hygiene, and the morning report 11:16 - Hill climbing a feature from a few feedback dumps 13:47 - Training a classifier overnight 15:27 - Pushing task scope, and the scaffolding it needs 17:06 - Never add reviewers to code you have not read canonical: >- https://liquidgold.cc/channel/aidotengineer/explore/video/prototyping-as-leadership-how-a-cto-ships-with-ai-a-d4aa10f4c69c markdown: >- https://liquidgold.cc/channel/aidotengineer/video/prototyping-as-leadership-how-a-cto-ships-with-ai-a-d4aa10f4c69c.md llms: null parent: 'https://liquidgold.cc/channel/aidotengineer.md' channel: aidotengineer updated: '2026-10-24T05:01:55.916Z' language: null children: [] videoId: bdHaOXZOhcM durationSeconds: 1098 publishedAt: '2026-08-20T14:00:21.000Z' searchChunks: - ordinal: 0 text: null startSeconds: 24 endSeconds: 170 watch: 'https://liquidgold.cc/watch/bdHaOXZOhcM?t=24' heldReason: rights - ordinal: 1 text: null startSeconds: 24 endSeconds: 170 watch: 'https://liquidgold.cc/watch/bdHaOXZOhcM?t=24' heldReason: rights - ordinal: 2 text: null startSeconds: 24 endSeconds: 170 watch: 'https://liquidgold.cc/watch/bdHaOXZOhcM?t=24' heldReason: rights - ordinal: 3 text: null startSeconds: 181 endSeconds: 268 watch: 'https://liquidgold.cc/watch/bdHaOXZOhcM?t=181' heldReason: rights - ordinal: 4 text: null startSeconds: 181 endSeconds: 268 watch: 'https://liquidgold.cc/watch/bdHaOXZOhcM?t=181' heldReason: rights - ordinal: 5 text: null startSeconds: 268 endSeconds: 279 watch: 'https://liquidgold.cc/watch/bdHaOXZOhcM?t=268' heldReason: rights - ordinal: 6 text: null startSeconds: 279 endSeconds: 324 watch: 'https://liquidgold.cc/watch/bdHaOXZOhcM?t=279' heldReason: rights - ordinal: 7 text: null startSeconds: 335 endSeconds: 366 watch: 'https://liquidgold.cc/watch/bdHaOXZOhcM?t=335' heldReason: rights - ordinal: 8 text: null startSeconds: 366 endSeconds: 378 watch: 'https://liquidgold.cc/watch/bdHaOXZOhcM?t=366' heldReason: rights - ordinal: 9 text: null startSeconds: 378 endSeconds: 421 watch: 'https://liquidgold.cc/watch/bdHaOXZOhcM?t=378' heldReason: rights - ordinal: 10 text: null startSeconds: 432 endSeconds: 577 watch: 'https://liquidgold.cc/watch/bdHaOXZOhcM?t=432' heldReason: rights - ordinal: 11 text: null startSeconds: 432 endSeconds: 577 watch: 'https://liquidgold.cc/watch/bdHaOXZOhcM?t=432' heldReason: rights - ordinal: 12 text: null startSeconds: 577 endSeconds: 588 watch: 'https://liquidgold.cc/watch/bdHaOXZOhcM?t=577' heldReason: rights - ordinal: 13 text: null startSeconds: 588 endSeconds: 684 watch: 'https://liquidgold.cc/watch/bdHaOXZOhcM?t=588' heldReason: rights - ordinal: 14 text: null startSeconds: 684 endSeconds: 694 watch: 'https://liquidgold.cc/watch/bdHaOXZOhcM?t=684' heldReason: rights - ordinal: 15 text: null startSeconds: 694 endSeconds: 836 watch: 'https://liquidgold.cc/watch/bdHaOXZOhcM?t=694' heldReason: rights - ordinal: 16 text: null startSeconds: 836 endSeconds: 847 watch: 'https://liquidgold.cc/watch/bdHaOXZOhcM?t=836' heldReason: rights - ordinal: 17 text: null startSeconds: 847 endSeconds: 937 watch: 'https://liquidgold.cc/watch/bdHaOXZOhcM?t=847' heldReason: rights - ordinal: 18 text: null startSeconds: 847 endSeconds: 937 watch: 'https://liquidgold.cc/watch/bdHaOXZOhcM?t=847' heldReason: rights - ordinal: 19 text: null startSeconds: 937 endSeconds: 947 watch: 'https://liquidgold.cc/watch/bdHaOXZOhcM?t=937' heldReason: rights - ordinal: 20 text: null startSeconds: 947 endSeconds: 1033 watch: 'https://liquidgold.cc/watch/bdHaOXZOhcM?t=947' heldReason: rights - ordinal: 21 text: null startSeconds: 1033 endSeconds: 1044 watch: 'https://liquidgold.cc/watch/bdHaOXZOhcM?t=1033' heldReason: rights --- # Prototyping as Leadership: How a CTO Ships with AI Agents — Hursh Agrawal, The Browser Company Fifteen recurring meetings a week, seven direct reports, a toddler at home, and somewhere inside that Hursh Agrawal ships between two and ten pull requests. It was not possible a year ago. His argument is that the manager schedule has quietly become usable as building time, which makes building part of a leadership job rather than a hobby squeezed around the edges of one. The reasoning is less about output than calibration. Frontier models turn over every few months and nobody can reliably tell you what a new one is good for, so a leader without hands on time ends up relaying opinions from Twitter. Building also solves a communication problem, since showing engineers a working prototype beats arguing about capability every quarter. He thinks leaders are unusually well suited to it as well, because they hold the most business context and their steering is therefore worth more per token than an engineer's. The mechanics come down to a single overnight loop. A coworker agent wired into Slack and the issue tracker spends twenty minutes assembling context into a prompt, that gets handed to a coding agent at five in the evening, and the morning block goes on reviewing whatever came back. He runs it to build features, to hill climb prompts against eval sets assembled from a handful of feedback dumps, and to train small classifiers end to end. The caveats are real. It only works on top of trustworthy CI, feature flags, and a prototype branch, and he is candid that his own code has caused incidents. Speaker info: - https://twitter.com/hursh - https://www.linkedin.com/in/hurshagrawal - https://www.hurshagrawal.com Timestamps: 0:00 - A manager's calendar, and why building became the job 2:44 - Judging a new model, and showing rather than arguing 4:25 - Execution is solved, judgment is not 5:16 - Four things leaders should build, and one to avoid 6:05 - The daily shape, and why 5pm matters most 6:56 - Building a feature overnight, and assembling the prompt 9:28 - Verification, PR hygiene, and the morning report 11:16 - Hill climbing a feature from a few feedback dumps 13:47 - Training a classifier overnight 15:27 - Pushing task scope, and the scaffolding it needs 17:06 - Never add reviewers to code you have not read Canonical: https://liquidgold.cc/channel/aidotengineer/explore/video/prototyping-as-leadership-how-a-cto-ships-with-ai-a-d4aa10f4c69c · Channel: [AI Engineer](https://liquidgold.cc/channel/aidotengineer.md) > Search chunk text is held until public-source provenance is verified. ## Document chunks > Missing: no stored document chunks are available.