Glean Introduces the Enterprise Agent Development Lifecycle, Codifying How Enterprises Build, Govern, and Measure AI Agents

New framework and platform capabilities help CIOs turn AI agents from fragmented experiments into governed production systems with measurable business impact

MOUNTAIN VIEW, Calif.--(BUSINESS WIRE)--Work AI leader Glean today introduced its enterprise Agent Development Lifecycle (ADLC), a new framework and set of platform capabilities designed to help enterprises systematically deploy AI agents and maximize business impact.

As organizations scale AI agents across teams, CIOs are under pressure to ensure those agents are useful, secure, and tied to business outcomes. But without a consistent, shared approach, enterprises risk exacerbating AI sprawl: agents scattered across teams, vendors, and workflows, with inconsistent governance and unclear ROI.

Glean’s answer is the ADLC, which gives CIOs and IT leaders a repeatable path for scaling agents across the business. The seven-stage lifecycle spans Opportunity, Design, Performance, Input, Develop, Launch, and Monitor & Improve - from identifying the business problem an agent should solve and designing the workflow, to defining success metrics, grounding the agent in enterprise context, building and testing it, launching with governance, and continuously improving it based on adoption, feedback, and business impact.

“At HubSpot, we’ve learned that successful agent adoption is not just about choosing the strongest model. It depends on giving AI the right enterprise context, creating structured enablement so employees know how to use it, and having a clear way to measure what is actually driving value,” said Rich Archbold, SVP Agentic GTM Engineering at HubSpot. “Glean has helped us bring those pieces together as we scale AI across HubSpot, giving our teams a trusted front door for building agents, accessing company context, and understanding where AI is delivering real impact.”

How Glean Brings the ADLC to Life

To bring the ADLC to life, Glean is introducing new capabilities and bringing together existing and upcoming platform investments across the stages where enterprises most often get stuck: building agents with the right context, launching them with the right governance, and measuring whether they are delivering value over time.

Build agents faster, with stronger context and better visibility

• Auto Mode Agent Builder: Users can describe what they want an agent to do in natural language, and the agent can plan, reason, and execute across the enterprise graph without predefined workflows or manual configuration.
• Debug & Trace Views: Full step-by-step visibility into every agent run, including inputs, tool calls, LLM decisions, and outputs, so builders can diagnose failures precisely rather than inferring from final output.
• Sub-Agents: Support for modular, production-grade agent architectures that allow parent agents to coordinate specialized agents at runtime.
• Expanded Agent Sandbox: Secure file system and code execution in the customer VPC, plus support for adding apps, not just individual actions.
• Content & Scheduled Triggers: Agents can react automatically to enterprise events such as content changes, scheduled runs, forms, and external events, allowing them to operate directly inside existing business processes.

Govern and distribute agents with greater control

• New Agent Library controls: Now generally available, verification badges, featured agents, departmental categories, and soft-delete with admin restore make the library a governed front door for agent distribution.
• Agent Access Policies: Organization-wide guardrails help enterprises apply consistent controls across agents, such as blocking or flagging sensitive content before an agent can process it, or restricting certain user groups from using agents to write to systems of record.

Treat agents like production systems

• Updated Agent Insights Dashboard: A rebuilt monitoring experience designed to track adoption, top use cases, estimated hours saved, and feedback trends over time, helping CIOs and builders understand which agents are delivering value and where continued improvement is needed.

“Enterprises spent the past year proving that agents can generate excitement. The next phase is proving they can generate results,” said Emrecan Dogan, Chief Product Officer at Glean. “Agents are software. They need a disciplined way to be defined, built, launched, governed, and improved over time. The Enterprise Agent Development Lifecycle gives CIOs a repeatable operating model for doing that, and Glean provides the platform capabilities to make it real.”

Availability:

• Generally Available: Auto Mode (including improved builder assistant), Debug & Trace Views, Agent Sandbox in auto mode agents, New Agent Library controls
• Beta: Content Triggers, Agent Access Policies
• Coming Soon: Updated Agent Insights Dashboard

Learn More:

• Register for today’s virtual Glean:LIVE event, taking place at 10:00 AM PT, to learn more
• Blog: Enable every agent to drive ROI with a robust agent development lifecycle
• eBook: The CIO’s Guide to Enterprise AI Agents

About Glean

Glean is the Work AI platform that helps everyone work smarter with AI. Glean Assistant gives every employee a powerful enterprise AI assistant that connects to and understands company data via Glean’s Enterprise Graph, and Glean Agents empowers everyone to create, use, and manage AI agents using natural language. Powered by Glean’s search and agentic engine, Glean’s agents automate work across the organization at scale, while ensuring permissions enforcement, full referenceability, governance, and security. By connecting across the enterprise technology ecosystem through native integrations, MCP servers, LLM choice, and APIs for customization, Glean delivers scalable, turnkey implementation of a complex AI ecosystem on one horizontal platform – without the need for costly professional services.