--- schema: 1 type: video title: How Kepler Built Verifiable AI for Financial Services — Vinoo Ganesh description: >- In finance a number is worthless until you can say where it came from. Vinoo Ganesh, CEO of Kepler, starts from the fact that language models are probability machines, brilliant at next token prediction and unreliable at the deterministic work, like arithmetic, that finance actually runs on. So Kepler treats the model as one part of a system rather than the whole answer, wrapping it in a deterministic substrate that makes every figure traceable. It rests on three tenets. Atomic provenance means every number is tied to its source and stripped out if it cannot be independently verified. Scope determinism keeps the model on the nondeterministic tasks it is good at and pulls the actual figure, say revenue from a 10K filing, deterministically behind the scenes, with reconciliation on top. The third is treating every extracted number like a pull request that gets reviewed, so entities are caught and nothing is invented. The result is a grounded system where the edge comes not from producing content but from verifying it. Speaker info: - https://x.com/vinooganesh - https://www.linkedin.com/in/vinoo-ganesh/ - https://vinoo.io Timestamps: 0:00 - Introduction: a data background in finance 1:42 - Why trust and verifiability matter now 2:57 - Models are probability machines 4:27 - Why analysts still put in the hours 8:22 - Modeling AI like an overworked VP 9:26 - Atomic provenance 12:01 - Scope determinism 13:52 - Reconciliation and pulling real numbers 15:06 - Extracting entities without misses 16:34 - Toward zero invented securities 20:31 - Where a number really comes from --- Watch along with the corrected transcript, summary, timestamps and resources on this talk's official AIE page: https://ai.engineer/talks/Tt2kX2sgQio-how-kepler-built-verifiable-ai-for-financial-services 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/how-kepler-built-verifiable-ai-for-financial-servic-befb5278d4c2 markdown: >- https://liquidgold.cc/channel/aidotengineer/video/how-kepler-built-verifiable-ai-for-financial-servic-befb5278d4c2.md llms: null parent: 'https://liquidgold.cc/channel/aidotengineer.md' channel: aidotengineer updated: '2026-10-24T06:29:44.865Z' language: null children: [] videoId: Tt2kX2sgQio durationSeconds: 1350 publishedAt: '2026-07-29T21:00:06.000Z' searchChunks: - ordinal: 0 text: null startSeconds: 112 endSeconds: 179 watch: 'https://liquidgold.cc/watch/Tt2kX2sgQio?t=112' heldReason: rights - ordinal: 1 text: null startSeconds: 112 endSeconds: 179 watch: 'https://liquidgold.cc/watch/Tt2kX2sgQio?t=112' heldReason: rights - ordinal: 2 text: null startSeconds: 191 endSeconds: 267 watch: 'https://liquidgold.cc/watch/Tt2kX2sgQio?t=191' heldReason: rights - ordinal: 3 text: null startSeconds: 191 endSeconds: 267 watch: 'https://liquidgold.cc/watch/Tt2kX2sgQio?t=191' heldReason: rights - ordinal: 4 text: null startSeconds: 277 endSeconds: 507 watch: 'https://liquidgold.cc/watch/Tt2kX2sgQio?t=277' heldReason: rights - ordinal: 5 text: null startSeconds: 277 endSeconds: 507 watch: 'https://liquidgold.cc/watch/Tt2kX2sgQio?t=277' heldReason: rights - ordinal: 6 text: null startSeconds: 277 endSeconds: 507 watch: 'https://liquidgold.cc/watch/Tt2kX2sgQio?t=277' heldReason: rights - ordinal: 7 text: null startSeconds: 519 endSeconds: 566 watch: 'https://liquidgold.cc/watch/Tt2kX2sgQio?t=519' heldReason: rights - ordinal: 8 text: null startSeconds: 578 endSeconds: 724 watch: 'https://liquidgold.cc/watch/Tt2kX2sgQio?t=578' heldReason: rights - ordinal: 9 text: null startSeconds: 578 endSeconds: 724 watch: 'https://liquidgold.cc/watch/Tt2kX2sgQio?t=578' heldReason: rights - ordinal: 10 text: null startSeconds: 734 endSeconds: 842 watch: 'https://liquidgold.cc/watch/Tt2kX2sgQio?t=734' heldReason: rights - ordinal: 11 text: null startSeconds: 734 endSeconds: 842 watch: 'https://liquidgold.cc/watch/Tt2kX2sgQio?t=734' heldReason: rights - ordinal: 12 text: null startSeconds: 852 endSeconds: 909 watch: 'https://liquidgold.cc/watch/Tt2kX2sgQio?t=852' heldReason: rights - ordinal: 13 text: null startSeconds: 909 endSeconds: 920 watch: 'https://liquidgold.cc/watch/Tt2kX2sgQio?t=909' heldReason: rights - ordinal: 14 text: null startSeconds: 920 endSeconds: 995 watch: 'https://liquidgold.cc/watch/Tt2kX2sgQio?t=920' heldReason: rights - ordinal: 15 text: null startSeconds: 920 endSeconds: 995 watch: 'https://liquidgold.cc/watch/Tt2kX2sgQio?t=920' heldReason: rights - ordinal: 16 text: null startSeconds: 1007 endSeconds: 1240 watch: 'https://liquidgold.cc/watch/Tt2kX2sgQio?t=1007' heldReason: rights - ordinal: 17 text: null startSeconds: 1007 endSeconds: 1240 watch: 'https://liquidgold.cc/watch/Tt2kX2sgQio?t=1007' heldReason: rights - ordinal: 18 text: null startSeconds: 1251 endSeconds: 1350 watch: 'https://liquidgold.cc/watch/Tt2kX2sgQio?t=1251' heldReason: rights - ordinal: 19 text: null startSeconds: 1251 endSeconds: 1350 watch: 'https://liquidgold.cc/watch/Tt2kX2sgQio?t=1251' heldReason: rights --- # How Kepler Built Verifiable AI for Financial Services — Vinoo Ganesh In finance a number is worthless until you can say where it came from. Vinoo Ganesh, CEO of Kepler, starts from the fact that language models are probability machines, brilliant at next token prediction and unreliable at the deterministic work, like arithmetic, that finance actually runs on. So Kepler treats the model as one part of a system rather than the whole answer, wrapping it in a deterministic substrate that makes every figure traceable. It rests on three tenets. Atomic provenance means every number is tied to its source and stripped out if it cannot be independently verified. Scope determinism keeps the model on the nondeterministic tasks it is good at and pulls the actual figure, say revenue from a 10K filing, deterministically behind the scenes, with reconciliation on top. The third is treating every extracted number like a pull request that gets reviewed, so entities are caught and nothing is invented. The result is a grounded system where the edge comes not from producing content but from verifying it. Speaker info: - https://x.com/vinooganesh - https://www.linkedin.com/in/vinoo-ganesh/ - https://vinoo.io Timestamps: 0:00 - Introduction: a data background in finance 1:42 - Why trust and verifiability matter now 2:57 - Models are probability machines 4:27 - Why analysts still put in the hours 8:22 - Modeling AI like an overworked VP 9:26 - Atomic provenance 12:01 - Scope determinism 13:52 - Reconciliation and pulling real numbers 15:06 - Extracting entities without misses 16:34 - Toward zero invented securities 20:31 - Where a number really comes from --- Watch along with the corrected transcript, summary, timestamps and resources on this talk's official AIE page: https://ai.engineer/talks/Tt2kX2sgQio-how-kepler-built-verifiable-ai-for-financial-services 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/how-kepler-built-verifiable-ai-for-financial-servic-befb5278d4c2 · 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.