--- schema: 1 type: video title: 'The Half Life of Agent Infrastructure — Ben Kus, Box' description: >- Ben Kus stood on this same stage last year and recommended a graph based approach to building agents. Someone found him afterward to say it was exactly the answer he needed. Kus opens this talk by retracting it, and admits he has since wondered what became of that person. Nothing about the advice was wrong when he gave it. A better way simply arrived, as it did for most of what was presented at that conference. He runs the same exercise across models, agent design and retrieval, and each chain lands somewhere unrecognizable from where it started twelve months earlier. The number that reframes it: infrastructure normally has a half life of three to five years, which is why the standard advice is to pick a stack, go deep and switch rarely, since every migration breaks something. Agent infrastructure has a half life measured in months. That breaks the usual advice, and it lands hardest on people. He tells it against himself, having asked an engineer to rebuild working agentic search in a new style, then proposed rebuilding it again the day after shipping. His answer is not a technology, it is preparation. Tell teams that change is normal rather than a failure, build abstractions that let the layer underneath be swapped, and only switch when eval sets say it matters, not when a paper is exciting. Box now reviews every AI technology on a six month clock. Speaker info: - https://x.com/benatbox - https://www.linkedin.com/in/benkus/ Timestamps: 0:00 - An exabyte, and now a trillion tokens 2:27 - The advice that used to work 4:48 - Retracting last year's talk from this stage 5:57 - How the model choices kept moving 7:07 - Agent design, from single calls to harnesses 8:15 - Retrieval, from keywords back to agents 10:31 - What does not change this fast 11:37 - Who a months long half life hurts 13:49 - Rebuilding it the day after shipping 14:57 - Telling teams that change is not a mistake 16:04 - Changing on evals, not on trends 17:13 - Judging vendors by how they handled change --- Watch along with the corrected transcript, summary, timestamps and resources on this talk's official AIE page: https://ai.engineer/talks/sM1iYgz93HI-youtube-b006941a973d92c44b1c Join our upcoming AI Engineer conferences (see https://ai.engineer landing page): - AIEi Shanghai, November 5–6 — the first for meeting top Chinese labs and engineers - AIE CODE in SF, November 10–12 — the top AI Coding conference returns - AIEi Sydney, December 7–8 — the first AIEi run alongside NeurIPS 2026 in Sydney --- canonical: >- https://liquidgold.cc/channel/aidotengineer/explore/video/the-half-life-of-agent-infrastructure-ben-kus-box-6c8f662b3fae markdown: >- https://liquidgold.cc/channel/aidotengineer/video/the-half-life-of-agent-infrastructure-ben-kus-box-6c8f662b3fae.md llms: null parent: 'https://liquidgold.cc/channel/aidotengineer.md' channel: aidotengineer updated: '2026-10-24T04:17:25.007Z' language: null children: [] videoId: sM1iYgz93HI durationSeconds: 1166 publishedAt: '2026-08-29T14:30:22.000Z' searchChunks: - ordinal: 0 text: null startSeconds: 162 endSeconds: 297 watch: 'https://liquidgold.cc/watch/sM1iYgz93HI?t=162' heldReason: rights - ordinal: 1 text: null startSeconds: 162 endSeconds: 297 watch: 'https://liquidgold.cc/watch/sM1iYgz93HI?t=162' heldReason: rights - ordinal: 2 text: null startSeconds: 376 endSeconds: 432 watch: 'https://liquidgold.cc/watch/sM1iYgz93HI?t=376' heldReason: rights - ordinal: 3 text: null startSeconds: 376 endSeconds: 432 watch: 'https://liquidgold.cc/watch/sM1iYgz93HI?t=376' heldReason: rights - ordinal: 4 text: null startSeconds: 432 endSeconds: 443 watch: 'https://liquidgold.cc/watch/sM1iYgz93HI?t=432' heldReason: rights - ordinal: 5 text: null startSeconds: 443 endSeconds: 497 watch: 'https://liquidgold.cc/watch/sM1iYgz93HI?t=443' heldReason: rights - ordinal: 6 text: null startSeconds: 443 endSeconds: 497 watch: 'https://liquidgold.cc/watch/sM1iYgz93HI?t=443' heldReason: rights - ordinal: 7 text: null startSeconds: 443 endSeconds: 497 watch: 'https://liquidgold.cc/watch/sM1iYgz93HI?t=443' heldReason: rights - ordinal: 8 text: null startSeconds: 497 endSeconds: 507 watch: 'https://liquidgold.cc/watch/sM1iYgz93HI?t=497' heldReason: rights - ordinal: 9 text: null startSeconds: 497 endSeconds: 507 watch: 'https://liquidgold.cc/watch/sM1iYgz93HI?t=497' heldReason: rights - ordinal: 10 text: null startSeconds: 507 endSeconds: 639 watch: 'https://liquidgold.cc/watch/sM1iYgz93HI?t=507' heldReason: rights - ordinal: 11 text: null startSeconds: 650 endSeconds: 704 watch: 'https://liquidgold.cc/watch/sM1iYgz93HI?t=650' heldReason: rights - ordinal: 12 text: null startSeconds: 650 endSeconds: 704 watch: 'https://liquidgold.cc/watch/sM1iYgz93HI?t=650' heldReason: rights - ordinal: 13 text: null startSeconds: 846 endSeconds: 902 watch: 'https://liquidgold.cc/watch/sM1iYgz93HI?t=846' heldReason: rights - ordinal: 14 text: null startSeconds: 902 endSeconds: 913 watch: 'https://liquidgold.cc/watch/sM1iYgz93HI?t=902' heldReason: rights - ordinal: 15 text: null startSeconds: 969 endSeconds: 981 watch: 'https://liquidgold.cc/watch/sM1iYgz93HI?t=969' heldReason: rights - ordinal: 16 text: null startSeconds: 1047 endSeconds: 1166 watch: 'https://liquidgold.cc/watch/sM1iYgz93HI?t=1047' heldReason: rights - ordinal: 17 text: null startSeconds: 1047 endSeconds: 1166 watch: 'https://liquidgold.cc/watch/sM1iYgz93HI?t=1047' heldReason: rights --- # The Half Life of Agent Infrastructure — Ben Kus, Box Ben Kus stood on this same stage last year and recommended a graph based approach to building agents. Someone found him afterward to say it was exactly the answer he needed. Kus opens this talk by retracting it, and admits he has since wondered what became of that person. Nothing about the advice was wrong when he gave it. A better way simply arrived, as it did for most of what was presented at that conference. He runs the same exercise across models, agent design and retrieval, and each chain lands somewhere unrecognizable from where it started twelve months earlier. The number that reframes it: infrastructure normally has a half life of three to five years, which is why the standard advice is to pick a stack, go deep and switch rarely, since every migration breaks something. Agent infrastructure has a half life measured in months. That breaks the usual advice, and it lands hardest on people. He tells it against himself, having asked an engineer to rebuild working agentic search in a new style, then proposed rebuilding it again the day after shipping. His answer is not a technology, it is preparation. Tell teams that change is normal rather than a failure, build abstractions that let the layer underneath be swapped, and only switch when eval sets say it matters, not when a paper is exciting. Box now reviews every AI technology on a six month clock. Speaker info: - https://x.com/benatbox - https://www.linkedin.com/in/benkus/ Timestamps: 0:00 - An exabyte, and now a trillion tokens 2:27 - The advice that used to work 4:48 - Retracting last year's talk from this stage 5:57 - How the model choices kept moving 7:07 - Agent design, from single calls to harnesses 8:15 - Retrieval, from keywords back to agents 10:31 - What does not change this fast 11:37 - Who a months long half life hurts 13:49 - Rebuilding it the day after shipping 14:57 - Telling teams that change is not a mistake 16:04 - Changing on evals, not on trends 17:13 - Judging vendors by how they handled change --- Watch along with the corrected transcript, summary, timestamps and resources on this talk's official AIE page: https://ai.engineer/talks/sM1iYgz93HI-youtube-b006941a973d92c44b1c Join our upcoming AI Engineer conferences (see https://ai.engineer landing page): - AIEi Shanghai, November 5–6 — the first for meeting top Chinese labs and engineers - AIE CODE in SF, November 10–12 — the top AI Coding conference returns - AIEi Sydney, December 7–8 — the first AIEi run alongside NeurIPS 2026 in Sydney --- Canonical: https://liquidgold.cc/channel/aidotengineer/explore/video/the-half-life-of-agent-infrastructure-ben-kus-box-6c8f662b3fae · 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.