Databricks Genie

The Production AI Playbook: Deploying Agents at Enterprise Scale — Sandipan Bhaumik, Databricks

Jun 18, 2026 · 37:06

Sandipan Bhaumik, a technical lead for Data and AI at Databricks, presents a five‑pillar playbook for taking AI agents to production: evaluation (define numerical success before touching code), observability (trace every decision for regulators and debugging), data foundation (agents do not forgive bad data), multi‑agent orchestration patterns (orchestrator‑worker, choreography, human‑in‑the‑loop), and governance (PII pre‑validation, prompt versioning as change management). He recounts a retail bank that spent £85,000 over six months on a chatbot PoC that failed because no one could measure or trace it. His team reversed the order: they built the evaluation dataset and tracing infrastructure first, selected the model in week 7 of an 8‑week engagement, and launched successfully. Six weeks post‑launch, when the bank updated interest rate policies, the tracing system caught that the new document had not been re‑embedded, so the agent served stale answers—a production incident the five pillars were designed to handle. The evaluation dataset is a living system that grows from 200 test cases; a production incident playbook connects all pillars: detect via eval dashboard, diagnose with…

Small Bets, Big Impact Building GenBI at a Fortune 100 – Asaf Bord, Northwestern Mutual

Dec 23, 2025 · 22:50

Asaf Bord, AI Product Lead at Northwestern Mutual, shares how his team built GenBI, an LLM-powered analytics copilot, by flipping the logic from a single big bet to an incremental roadmap of small, fundable projects. Using real, messy data from the 160-year-old company, they deployed a modular architecture with metadata, RAG, SQL, and BI agents, each productizable independently. The RAG agent alone automated 80% of the 20% of BI team capacity spent on finding and sharing reports, saving roughly two full-time employees. Bord explains how a crawl-walk-run release strategy built trust with both users and leadership, starting with BI experts before expanding to business managers, and how each six-week sprint delivered tangible business value—like proving the ROI of enriched metadata through A/B tests against a semantic layer initiative. He also explores the future of SaaS pricing in the GenAI era, questioning whether per-seat models still make sense when individuals become 10x more effective.