Today at Glean:GO, something meaningful happened.

We’ve all been hearing the excitement around AI agents — copilots, assistants, automation bots — but if we’re honest, most of it has felt like early-stage experimentation. A lot of promise. Some solid use cases. But few examples of true enterprise-scale readiness.

That changed today.

Glean announced the general availability of Glean Agents, and it’s one of the most complete, enterprise-ready frameworks I’ve seen to bring agents into production — across people, workflows, and systems.

I've had a front-row seat to how enterprises are thinking about AI. I get to speak with executives and builders every week. The questions are always the same:

• How do we deploy AI agents at scale?

• How do we keep control of data, cost, and risk?

• How do we go beyond pilots and make this real?

Today, Glean answered these questions with clarity.

What Glean Announced Today — and Why It Matters

Glean Agents is not just a feature. It’s a platform—a horizontal, open environment for building, deploying, orchestrating, and governing AI agents that actually move the business forward.

And unlike many others, Glean has built this with real enterprise guardrails and architecture in mind.

Let’s break it down.

1. Build Agents, Your Way

Whether you're a business user or a developer, Glean Agents meets you where you are:

• Natural Language Agent Builder: Describe what you want in plain language, and Glean will generate a functional agent for you — no technical expertise needed.

• Agents API: Developers can define logic, build agent flows programmatically, and coordinate across systems and models.

• Agent Library: Agents are reusable and shareable across teams, helping businesses scale what's working.

This blend of simplicity and control is exactly what enterprises need.

2. Freedom to Choose Models

One of the standout features for me is Glean’s per-step model selection. You’re not locked into one LLM. You can tailor each stage of an agent’s logic to use a model based on speed, quality, or cost — whether that’s OpenAI, Claude, Vertex AI, or Bedrock.

This flexibility is huge.

And the universal model key? A single layer of access across 15+ leading LLMs — without needing to negotiate with each provider separately. That’s a game-changer for scaling responsibly.

3. Agents That Collaborate and Coordinate

Agents that work in isolation are interesting. But agents that collaborate are powerful.

With their new hosted Model Context Protocol (MCP) server, Glean enables agent-to-agent communication — across platforms. One agent can trigger another, share data, and work together toward a shared goal.

Glean also announced a strategic partnership with Workday, enabling joint workflows across HR and finance. That’s not just integration — it’s orchestration.

4. Prebuilt Agents Ready to Use

You don’t need to start from scratch. Glean launched:

• 30+ prebuilt Quickstart agents across sales, IT, HR, and engineering — from prospecting to help desk automation.

• Personal productivity agents for meeting prep, action summaries, and performance reviews — powered by your personal graph.

• Deep research agents that combine internal and external data with citations and analysis.

• Structured query agents for Snowflake Cortex — enabling real-time, natural language querying of your enterprise data.

This is a practical jumpstart — and a signal that Glean isn’t just shipping infrastructure. They’re focused on outcomes.

5. Enterprise-Grade Security and Observability

Security isn’t optional. And Glean doesn’t treat it like an afterthought.

Built-in protections include:

• Prompt injection and jailbreak defenses

• Sensitive content detection across 100+ enterprise systems

• Granular access controls by user and role

• Integrated observability and evals — so teams can trace agent behavior step-by-step and improve outcomes continuously

This brings trust to the conversation. It’s how you move from pilot to production with confidence.

6. On-Premise Support — Powered by Dell

One of the most enterprise-savvy moves announced today was Glean’s partnership with Dell Technologies.

For organizations that can't rely solely on cloud due to regulatory or compliance constraints, Glean is now available on-prem, running on Dell’s trusted infrastructure. This unlocks secure deployment of Glean Agents and enterprise search in private data centers.

If you’ve worked with healthcare, financial services, or government clients — you know how important this option is.

My Reflections from the Show Floor

It’s easy to get excited about AI. But what impressed me most at Glean:GO wasn’t just the product. It was the level of intentionality in how it was built.

Glean has taken some of the toughest problems in Work AI — interoperability, model diversity, data governance, user control — and solved them in a platform that’s flexible, usable, and secure.

This isn’t theory. It’s available now.

They’ve also backed it up with meaningful partnerships — with Workday, Snowflake, Dell, and Palo Alto Networks — to give enterprises not just software, but an ecosystem to scale with.

For me, this felt like the day enterprise AI agents turned the corner — from proof of concept to platform. From vision to deployment.

If You’re an Enterprise Thinking About Agents

Ask yourself:

• Are your agents grounded in your enterprise knowledge?

• Can your teams build and iterate without deep technical expertise?

• Do your agents collaborate, or do they live in silos?

• Do you have governance in place?

• And are you seeing value yet?

If the answer to any of these is “not yet,” this might be the moment to look again.

Today’s announcement from Glean is a big signal that the agentic future is arriving — not in theory, but in practice.

I’ll be breaking down more of these trends and use cases on The Ravit Show in the coming weeks.

Let’s keep the conversation going.

Until next time,
Ravit Jain
Founder & Host, The Ravit Show