Launch Week Day 5: SOC2 Type 2 Compliance

Today marks the final day of our first Launch Week, and we're wrapping things up with an announcement that might not be the flashiest, but is arguably one of the most important for teams building production AI applications: Agenta is now SOC2 Type 2 compliant.

What This Actually Means

For those who've been through security reviews before, you know the drill. For everyone else, SOC2 Type 2 is essentially independent verification that we've implemented robust security controls and have been consistently following them over time.

While SOC2 Type 1 (which we announced previously) was a point-in-time audit of our security controls, Type 2 is the more rigorous version that verifies we've been sticking to these controls over a sustained period. This matters for several reasons:

• Enterprise readiness: If you're selling to larger companies, this often removes a major obstacle

• Data security: Verification that we're handling your prompts, configurations, and observability data with proper controls

• Operational maturity: Independent validation that our internal processes meet industry standards

We know security certifications aren't the most thrilling announcement, but for many of you building AI applications for regulated industries or enterprise clients, this removes a significant barrier to using Agenta in production.

A Quick Review of Launch Week

We've packed a lot into this week! In case you missed any of the announcements, here's a quick recap:

Day 1: AI Model Hub

We introduced the AI Model Hub, giving you access to a wide range of LLM providers and custom models directly within Agenta. This means you can now test, compare, and deploy with almost any model without leaving the platform.

Day 2: Custom Workflows

We launched support for complex workflows, including Agents. This allows you to build, test and evaluate sophisticated multi-step AI pipelines and agent architectures through our playground and evaluation framework.

Day 3: Prompt & Configuration Registry

We unveiled our comprehensive prompt registry with built-in versioning, giving teams a central repository to track, manage, and collaborate on prompts across projects.

Day 4: Structured Output in the Playground

We added native support for structured output in the playground, making it dramatically easier to work with JSON, parse responses, and build applications that need consistent, schema-validated outputs.

Day 5: SOC2 Type 2 Compliance (Today)

As discussed above, we've achieved SOC2 Type 2 compliance, making Agenta suitable for teams working with sensitive data or in regulated environments.

What's Next?

Launch Week might be ending, but we're just getting started. We've got a packed roadmap ahead, with a particular focus on improving the evaluation workflow in Agenta.

Thanks for following along all week. We've been blown away by the community feedback, excitement, and the creative ways everyone is already putting these new features to use. We're building Agenta for developers like you, so keep that feedback coming.

This won't be our last Launch Week either—we're already planning the next one with even more ambitious features. Join our Slack and star our repository to stay updated.