--- schema: 1 type: video title: >- Paving the Road to Production: Governed Enterprise AI Agents — Harish Gaggar, Credit Karma description: >- Register an agent through an internal portal and receive more than a repository: the deployment also needs identity, memory, traces, evaluations and secret handling. Harish Gaggar walks through that paved path for enterprise agents at Intuit Credit Karma. Local development progresses through container builds, review, security scanning and Kubernetes deployment, with the supporting resources provisioned consistently. The architecture places a centralized MCP gateway between agents and tools so authentication and routing are part of every request. Custom MCP servers expose reviewed tools, while OpenTelemetry, Tempo, Prometheus and Grafana make failures, latency and access behavior visible. Gaggar then follows a LangGraph workflow from interpretation and planning through evidence gathering, policy checks and human approval for sensitive actions. Threads, runs and checkpoints preserve state, while separate tracking tables record model use, tool calls, costs and evaluation fields. Demonstrations inspect those records rather than establish a validated quality score for the whole system. A PII filtering layer aims to intercept sensitive inputs before they reach models, memory or traces; detection and redaction remain controls to evaluate, rather than a guarantee that no secret can escape. The final component prunes, reranks, deduplicates and compresses context. Its examples show large token reductions, including about 90% in a demonstrated query, without a paired answer quality study or measured business ROI. The practical point is to build the surrounding framework from the start, giving each agent a consistent route through access control, observability, durable state and accountable action. Speaker info: - https://www.evidyafoundation.com/about-us Timestamps: 00:00 - From experiments to enterprise agents 1:28 - Missing production foundations 2:54 - The agent and supporting frameworks 4:06 - Registration, local development and the paved path 6:27 - Review, containers and Kubernetes deployment 8:14 - Reviewed custom MCP servers 9:47 - MCP gateway authentication and routing 11:48 - Logs, traces and gateway metrics 14:01 - LangGraph policies and human approval 16:59 - Threads, runs and memory checkpoints 19:48 - Agent tracking and evaluation records 23:27 - Filtering sensitive inputs 24:25 - Context optimization and token reduction demos 27:33 - Build the framework around every agent --- Watch along with the corrected transcript, summary, timestamps and resources on this talk's official AIE page: https://ai.engineer/talks/tZycGA_Fbfo-youtube-86707775be4ad567144b 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/watch/tZycGA_Fbfo' markdown: 'https://liquidgold.cc/watch/tZycGA_Fbfo.md' llms: 'https://liquidgold.cc/watch/tZycGA_Fbfo/llms.txt' parent: 'https://liquidgold.cc/channel/aidotengineer.md' channel: aidotengineer updated: '2026-10-23T22:23:59.630Z' language: null children: [] videoId: tZycGA_Fbfo durationSeconds: 1767 publishedAt: '2026-10-14T17:00:04.000Z' searchChunks: - ordinal: 0 text: null startSeconds: 104 endSeconds: 180 watch: 'https://liquidgold.cc/watch/tZycGA_Fbfo?t=104' heldReason: rights - ordinal: 1 text: null startSeconds: 180 endSeconds: 191 watch: 'https://liquidgold.cc/watch/tZycGA_Fbfo?t=180' heldReason: rights - ordinal: 2 text: null startSeconds: 191 endSeconds: 247 watch: 'https://liquidgold.cc/watch/tZycGA_Fbfo?t=191' heldReason: rights - ordinal: 3 text: null startSeconds: 257 endSeconds: 387 watch: 'https://liquidgold.cc/watch/tZycGA_Fbfo?t=257' heldReason: rights - ordinal: 4 text: null startSeconds: 398 endSeconds: 498 watch: 'https://liquidgold.cc/watch/tZycGA_Fbfo?t=398' heldReason: rights - ordinal: 5 text: null startSeconds: 398 endSeconds: 498 watch: 'https://liquidgold.cc/watch/tZycGA_Fbfo?t=398' heldReason: rights - ordinal: 6 text: null startSeconds: 602 endSeconds: 712 watch: 'https://liquidgold.cc/watch/tZycGA_Fbfo?t=602' heldReason: rights - ordinal: 7 text: null startSeconds: 723 endSeconds: 844 watch: 'https://liquidgold.cc/watch/tZycGA_Fbfo?t=723' heldReason: rights - ordinal: 8 text: null startSeconds: 855 endSeconds: 1020 watch: 'https://liquidgold.cc/watch/tZycGA_Fbfo?t=855' heldReason: rights - ordinal: 9 text: null startSeconds: 1032 endSeconds: 1188 watch: 'https://liquidgold.cc/watch/tZycGA_Fbfo?t=1032' heldReason: rights - ordinal: 10 text: null startSeconds: 1200 endSeconds: 1413 watch: 'https://liquidgold.cc/watch/tZycGA_Fbfo?t=1200' heldReason: rights - ordinal: 11 text: null startSeconds: 1423 endSeconds: 1468 watch: 'https://liquidgold.cc/watch/tZycGA_Fbfo?t=1423' heldReason: rights - ordinal: 12 text: null startSeconds: 1479 endSeconds: 1654 watch: 'https://liquidgold.cc/watch/tZycGA_Fbfo?t=1479' heldReason: rights - ordinal: 13 text: null startSeconds: 1479 endSeconds: 1654 watch: 'https://liquidgold.cc/watch/tZycGA_Fbfo?t=1479' heldReason: rights - ordinal: 14 text: null startSeconds: 1664 endSeconds: 1767 watch: 'https://liquidgold.cc/watch/tZycGA_Fbfo?t=1664' heldReason: rights --- # Paving the Road to Production: Governed Enterprise AI Agents — Harish Gaggar, Credit Karma Register an agent through an internal portal and receive more than a repository: the deployment also needs identity, memory, traces, evaluations and secret handling. Harish Gaggar walks through that paved path for enterprise agents at Intuit Credit Karma. Local development progresses through container builds, review, security scanning and Kubernetes deployment, with the supporting resources provisioned consistently. The architecture places a centralized MCP gateway between agents and tools so authentication and routing are part of every request. Custom MCP servers expose reviewed tools, while OpenTelemetry, Tempo, Prometheus and Grafana make failures, latency and access behavior visible. Gaggar then follows a LangGraph workflow from interpretation and planning through evidence gathering, policy checks and human approval for sensitive actions. Threads, runs and checkpoints preserve state, while separate tracking tables record model use, tool calls, costs and evaluation fields. Demonstrations inspect those records rather than establish a validated quality score for the whole system. A PII filtering layer aims to intercept sensitive inputs before they reach models, memory or traces; detection and redaction remain controls to evaluate, rather than a guarantee that no secret can escape. The final component prunes, reranks, deduplicates and compresses context. Its examples show large token reductions, including about 90% in a demonstrated query, without a paired answer quality study or measured business ROI. The practical point is to build the surrounding framework from the start, giving each agent a consistent route through access control, observability, durable state and accountable action. Speaker info: - https://www.evidyafoundation.com/about-us Timestamps: 00:00 - From experiments to enterprise agents 1:28 - Missing production foundations 2:54 - The agent and supporting frameworks 4:06 - Registration, local development and the paved path 6:27 - Review, containers and Kubernetes deployment 8:14 - Reviewed custom MCP servers 9:47 - MCP gateway authentication and routing 11:48 - Logs, traces and gateway metrics 14:01 - LangGraph policies and human approval 16:59 - Threads, runs and memory checkpoints 19:48 - Agent tracking and evaluation records 23:27 - Filtering sensitive inputs 24:25 - Context optimization and token reduction demos 27:33 - Build the framework around every agent --- Watch along with the corrected transcript, summary, timestamps and resources on this talk's official AIE page: https://ai.engineer/talks/tZycGA_Fbfo-youtube-86707775be4ad567144b 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/watch/tZycGA_Fbfo · Channel: [AI Engineer](https://liquidgold.cc/channel/aidotengineer.md) > Search chunk text is held until public-source provenance is verified. ## Document chunks --- #### Chunk 0: Document excerpt (0:03 to 1:32) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 1: Document excerpt (1:32 to 3:00) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 2: Document excerpt (3:00 to 4:17) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 3: Document excerpt (4:17 to 6:27) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 4: Document excerpt (6:27 to 8:18) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 5: Document excerpt (8:18 to 9:51) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 6: Document excerpt (9:51 to 12:03) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 7: Document excerpt (12:03 to 14:04) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 8: Document excerpt (14:04 to 17:00) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 9: Document excerpt (17:00 to 19:48) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 10: Document excerpt (19:48 to 23:33) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 11: Document excerpt (23:33 to 24:28) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 12: Document excerpt (24:28 to 27:34) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 13: Document excerpt (27:34 to 29:27) > Held: transcript rights do not permit this chunk's text.