01 — Executive Summary
We went in as a customer. We left as a reference.
Summit confirmed that RBNA's Wingman initiative is not just aligned with where Snowflake is heading — our work is actively shaping their product roadmap. Multiple Snowflake PMs described our team's approach as ahead of the curve. Our investment in semantic views, governed agent deployment, and rigorous validation puts us in a position that most of Snowflake's enterprise customers are still working toward.
02 — Platform Direction
Where Snowflake is going — and why our bets are right
Snowflake's direction is unambiguous: the entire company is pivoting around the Agentic Enterprise — governed AI agents that don't just answer questions, but take action. Our current Wingman architecture anticipates this shift.
- AISnowflake Intelligence → CoWork (now GA): Evolved from Q&A agents to a personal work agent with user memory, skills, and multi-agent orchestration. Features we've been requesting — skill catalogs, mobile access, curated publishing — are now shipping.
- DATACortex Sense Runtime: Auto-builds business context with zero manual setup, achieving 86% accuracy on hard enterprise questions vs. 24% baseline. Directly addresses the trust gap we currently manage with manual QA review.
- SAPSAP-Snowflake integration now GA: Bidirectional semantic view pipeline with automated lineage and CSN conversion. Given our SAP dependency, this unlocks future data unification without pipeline duplication.
- ENGdbt on Snowflake — CI/CD modernization: Direct session with Uche (Snowflake PM, dbt Products) gave us a clear architecture: GitHub Actions + Snowflake CLI + zero-copy clone CI. Eliminates our single-developer deployment bottleneck.
- GOVHorizon governance expansion: New intent-driven governance, agent identity policies, and multi-party approvals give us the tools to scale Wingman without increasing manual oversight.
Strategic alignment
Snowflake's keynote framed the central enterprise challenge as: "85% of clients have a data problem." RBNA is past that. The industry is catching up to where we already are.
03 — PM Sessions & Our Position
What Snowflake PMs said — directly
Our team secured direct briefings with Snowflake PMs across Intelligence/CoWork, Observability, Semantic Layer, and dbt Products. Every session surfaced the same signal: RBNA is ahead of the curve, and our feedback is actively shaping the roadmap.
Effie Goenawan
Principal Product Manager · Intelligence & CoWork
Snowflake Intelligence & CoWork roadmap
A dedicated session with 9 Snowflake PMs — the largest PM engagement of the Summit. Confirmed RBNA is a top "Artifacts PRP" customer. Effie explicitly stated that Red Bull's interest in charts has shaped the charting roadmap, and that our conversations inform Snowflake's direction broadly. New dashboarding capabilities previewed: AI-authored dashboards via Coco, waterfall and box-whisker chart support, RBAC-controlled publishing, and conversational follow-up ("Coco talk") directly on published dashboards.
The memo that made the keynote: Sameer, Avi, and Neil submitted a memo with 12 feature requests for the Intelligence platform. Snowflake built 9 of the 12 — all showcased in the Summit keynote on June 2, 2026.
Also in this session
Alan Jin
Sr. PM · SQL & Workspaces
Shrikant Shanbhag
Director of Engineering
Alex Issa
Product Manager · AI Observability & Evaluations
Agent monitoring, observability & lifecycle
Validated that RBNA's command center approach is among the most structured feedback loops he sees from enterprise customers. Background in distributed systems, ML, and serverless architecture; CS degrees from UT Austin. Our methodology is being documented as a reference pattern within Snowflake's PM team.
Agent versioning now GA. Solves our multi-version testing clutter directly — tag versions with aliases (default, beta), split traffic between them, manage via Git-backed YAML in Agent Studio.
LLM-based automated monitoring (July private preview): replaces our unsustainable manual daily review. "LLM judges" auto-score traces on answer relevance, plan quality, logical consistency, and a new "user frustration" metric.
Coco Evolve (pre-research preview): a "human-out-of-the-loop" agent builder inspired by DeepMind's AlphaEvolve. Proof of concept recreated an agent that took 4 developers 4 months — in 6 hours.
Josh Klahr
Head of Product · AI-Powered BI
Semantic views, ontology & OSI alignment
25+ years in data and analytics — previously VP of Product at Splunk, AtScale, and Pivotal. Josh told us directly that our team is "ahead of the game" relative to most Snowflake customers. Our Python-based ontology-first approach mirrors how Cortex Sense internally represents its own knowledge graph — Josh recommended the "Ontology on Snowflake" Labs solution as the closest match to what we've already built. He requested our SAP BW nested aggregation examples to share with engineering for LOD evaluation. Three additional private previews unlocked: LOD support, Power BI PBIT model export (June), and XMLA endpoint exposing Snowflake semantic views as SSAS (end of June).
Uche Nkadi
Product Lead · dbt Projects on Snowflake
dbt on Snowflake — CI/CD & orchestration
Described our team as ahead of most customers in dbt maturity. Specializes in Developer Experience, Developer Tooling, and Observability. Recommended CI/CD architecture: GitHub Actions + Snowflake CLI + zero-copy clone CI. Fast-tracked RBNA for two private previews: VSCodium dbt IDE (June 15) and advanced CI/CD cloning.
"Red Bull's interest in charts has informed a lot of our roadmap on the charts — and in general, Snowflake's roadmap is informed from a lot of our conversations."
Effie Goenawan
Principal Product Manager, Snowflake Intelligence & CoWork · Direct PM session, Summit 2026
04 — What We Are Doing With What We Learned
Concrete actions & account expectations
Accessing New Features — Private Preview
Personal Work Agent PRP
Early access to user memory, skills, multi-agent orchestration, and automations — features we requested. Enrolling RBNA leaders ahead of general availability.
Enroll by June 8
dbt IDE Private Preview — VSCodium + Cortex
Snowflake-managed IDE with built-in Cortex AI panel. Eliminates manual workspace deployment dependency. Our slot confirmed by Uche.
June 15
Dashboards Private Preview
New Snowflake-native dashboarding — comparable to SAP Analytics Cloud and Power BI, built on Streamlit. AI-authored dashboards via Coco, advanced chart types (waterfall, box-whisker), RBAC-controlled publishing, and conversational follow-up directly on dashboards. Recommended to wait for mid-June build for the best experience.
June 15–17
Agent Studio — new agent builder experience
Git-backed YAML configuration, version aliases (default/beta), traffic splitting between versions, and built-in observability metrics (cost, latency, active users, error rates). Signed up via Alex Issa session. Directly solves our multi-version testing clutter.
Signing up now
TEA & TSA Eval Metrics — Public Preview
Tool Execution Accuracy and Tool Selection Accuracy — new evaluation metrics that measure whether agents are selecting and executing the right tools. Entering public preview week of June 9.
Week of June 9
Evals — Agent Version Targeting & Dataset Versioning
Ability to point evaluations at specific agent versions and version eval datasets as native Snowflake objects. Enables proper A/B testing of agent improvements without contaminating production evals.
Mid-June
LLM-based Automated Monitoring
Replaces our current unsustainable manual daily review process. "LLM judges" automatically score agent traces on answer relevance, plan quality, logical consistency, and user frustration — surfacing only what needs human attention.
July PrPr
Export Snowflake semantic views directly as Power BI models. Unlocked from our Josh session — gives teams a path to consume governed semantic layer data in Power BI without rebuilding models.
June
