TITLE: Snowflake Launches Agentic AI That Executes Work on Your Data DATE: 2026-03-21 COMPANY: Snowflake TOPIC: Agent Systems SUMMARY: Snowflake announced Project SnowWork on 18 March 2026, a new agentic AI platform that autonomously completes multi-step business workflows from plain-language prompts. Built on a company's own governed data, it handles tasks like pulling figures, building analysis, generating deliverables, and drafting follow-up communications without human hand-holding. The platform enters research preview with a limited set of customers and no disclosed pricing. WHAT CHANGED: On 18 March 2026, Snowflake announced the research preview of Project SnowWork, an agentic AI platform built to complete multi-step business workflows from plain-language instructions. A user can describe what they need, and the platform plans the required steps, retrieves governed data, runs analysis, synthesises insights, and generates finished deliverables, including reports, presentations, and follow-up communications, within a single interaction. Project SnowWork is built on Snowflake's enterprise data platform, meaning it operates on a company's actual figures rather than generic AI knowledge. It inherits Snowflake's existing role-based access controls, data masking policies, and audit logging, so the AI works within the same security boundaries as the data it touches. The platform includes pre-built, role-specific skill profiles for common business functions including finance, sales, marketing, and operations. These profiles are pre-configured with the workflows, terminology, and KPIs relevant to each function, reducing setup time for non-technical users. Sridhar Ramaswamy, Snowflake's CEO, described the launch as a step into "the era of the agentic enterprise," positioning Project SnowWork as the third pillar of Snowflake's AI stack alongside Snowflake Intelligence (natural language question-answering, now generally available) and Cortex Code (AI for data engineering and application development). WHY IT MATTERS: Agentic AI is crossing from developer tools into the hands of business users. Operators no longer need technical staff to unlock the value of automation. Building agents on governed enterprise data is a material advantage over general-purpose AI. Outputs are grounded in the organisation's own figures, not estimates or external proxies. Role-specific profiles mean teams can act within hours of deployment rather than weeks of configuration. Native governance and audit logging address one of the primary enterprise objections to AI agents: the risk of agents accessing data they should not. The "control plane" architecture Snowflake describes, which coordinates AI-driven actions across systems within defined policies, is the correct model for scaling agents without losing compliance. Project SnowWork signals that data platform vendors are moving aggressively into workflow automation, directly competing with traditional software tools. DAVID & GOLIATH ANALYSIS: Project SnowWork is worth watching closely because it solves a problem most AI tools ignore: finishing the job. The dominant pattern in enterprise AI today is augmented intelligence, tools that surface information faster and help humans make decisions. Project SnowWork is designed to take the next step and complete the deliverable without waiting for human assembly. For operators running lean teams, this distinction is consequential. A finance manager who can describe a reporting task in plain language and receive a finished, governed, audit-ready output is not just saving time. They are fundamentally changing how many people they need to run a particular function. That is the productivity geometry that matters for organisations competing with much larger enterprises. The limitation to note is access. Project SnowWork is in research preview with no pricing or timeline disclosed. It requires Snowflake as the underlying data platform, which is not the right fit for every organisation. Operators should note the pattern regardless: agentic tools that work on your own data, within your existing governance rules, are the category to prioritise in any AI evaluation this year. RELEVANT SYSTEMS: Employee Amplification Systems, AI Growth Engine, Secure AI Brain SOURCE URL: https://davidandgoliath.ai/daily-ai-briefing/snowflake-project-snowwork-agentic-ai-enterprise FEED URL: https://davidandgoliath.ai/daily-ai-briefing/feed --- Published by David & Goliath | https://davidandgoliath.ai Daily AI Briefing: one AI development per day, decoded for business operators. This is a structured companion file optimised for LLM retrieval and citation.