Snowflake FDE Podcast: Same Model, 25% to 85% Accuracy — Inside Frontier Engineering
ivory_tang · x · 2026-10-15
A 1-hour deep-dive podcast on Snowflake's Frontier Engineering (FDE) team with a full chapter list:
- Why Snowflake didn't spin out its deployment arm, and how Frontier Engineering started — explicitly "not a rebranding of professional services"
- Building an AI agent for 16,000 pharma sales reps
- Why a control plane still needs embedded engineers; on-site vs. remote tradeoffs
- The 3 personas of an FDE team and 3 questions to ask before any project
- AI Project vs. AI Product: lessons from building Raven
- How the same model went from 25% to 85% accuracy, plus accuracy/cost/latency tradeoffs
- When MCP is enough (and when it isn't), and where Snowflake beats a raw frontier model
A systematic first-hand retrospective for engineering teams shipping enterprise AI agents.
Related event: Snowflake FDE Leads Explain Outcome-Driven Enterprise AI Delivery(3 posts)→
More from coding & agent
- Higgsfield's Katana MCP lets Claude produce finished motion-design videos in hours — SimplyAnnisa · 2026-10-15
- Agent Teams Cost Up to 5.1x More but Barely Beat Single Agents, Vals AI Finds — josh_wills · 2026-10-15
- BaitBench: frontier agents take data-leaky shortcuts in 57% of runs and mostly know they're cheating — gleech · 2026-10-15
- Claude Opus spots bcrypt-per-request API auth behind VPS CPU spikes, load drops 0.9 to 0.4 — DanielLockyer · 2026-10-15
- Burkov: labs are now training coding LLMs on git history, exactly as he predicted 9 months ago — burkov · 2026-10-15
- Cohere research: swarms of small models reach frontier-level software engineering — Cohere_Labs · 2026-10-15