Our powerful new capabilities - from a no-code Studio Assistant to a full workflow process engine - make Vertesia easier to operate, and more powerful to build on.
Let me walk you through what's new, and what it means for your teams.
The Studio Assistant is a powerful agentic assistant embedded directly inside Vertesia Studio. It makes creating and using agents quicker and more productive.
Using the Studio Assistant, you just describe what you want. "I need an agent that monitors our competitor pricing every morning and sends me a summary." The Studio Assistant writes the prompts, tests with different models, learns from the execution, refines and publishes the agent. Previous to the release of the Studio Assistant, all of these steps had to be done manually. Now, they’re handled by our agentic assistant, making you more productive and Vertesia easier to use.
Our Studio Assistant isn’t limited to only agents. You can also use the Studio Assistant to manage content, types, collections, databases and dashboards.
The most powerful AI agent is one that doesn't wait for you to start it.
Our Agent Scheduler transforms conversational agents into proactive automation engines. You can configure any agent to run on a recurring schedule (daily, weekly, monthly, or any custom interval). And because agents can schedule themselves from within a conversation, the system is fully autonomous: an agent can decide mid-task that it needs to run a recurrent follow-up analysis every Monday at 9am and create that schedule without any human intervention.
What does this look like in practice? A financial services team running daily pre-market briefings with no manual trigger. A compliance team that gets weekly regulatory scan reports delivered automatically. A marketing team whose competitive analysis agent wakes up every Monday morning and has insights ready before the first standup.
These agents run like any other agent, but on a schedule. Meaning, agents run with your identity, permissions and access tools, along with full observability and analytics about their execution. You’ll always have visibility into and understanding of the cost and performance of these scheduled agents.
Not every system has an API. In fact, most of the web doesn't.
Government portals. Legacy vendor platforms. These systems aren't going away - and in a world where AI agents are supposed to handle end-to-end workflows, hitting a wall at a web form with no integration breaks the entire promise.
Agentic browsing closes that gap. It reads and understands web pages the way a human does: semantically, adaptively, in real time. Agentic browsing runs entirely in the cloud, isolated from your local environment, with enterprise-grade authentication and no installation required.
We tested it against live websites, and here’s what we got: An agent configured a 2027 BMW M5 on bmwusa.com, navigating a live configurator across multiple sessions, adapting when options changed, and returning a complete specification with pricing. Another agent worked through Google Trends - a site with no API and a fully dynamic JavaScript interface - across 48 browser iterations, then filtered out contextually irrelevant results before synthesizing a ranked keyword strategy. Crucially, these agents aren't just browsing public websites; they can securely authenticate on behalf of the user, navigating multi-factor authentication (MFA) seamlessly to access walled gardens. While we view agentic browsing as a powerful 'last resort' when MCPS or APIs aren't available, these tests prove that when the guardrails are down, the agents can handle the messiest parts of the web.
The web was built for humans. Now your agents can navigate it too - safely, intelligently, and without breaking every time the UI changes.
And for teams building complex, multi-step business processes, we're introducing the Process Engine. This is our answer to the enterprise AI adoption gap: the productive middle ground between fully autonomous agents (powerful but not always necessary) and rigid deterministic automation (predictable but inflexible).
The Process Engine helps you define and build workflow processes from a diagram, documentation or an agent conversation.
Take an image of your workflow process diagram and upload it into Vertesia. The process engine then builds an interactive workflow for your process. Next, let an agent run through all of the individual steps and possible combinations to thoroughly test the workflow before publishing. This works also for process documentation, technical configurations (like BPEL), or a conversation with the agent about the workflow process.
Think loan origination. Contract review. Prior authorization. RFP evaluation. These are the workflows that matter most to regulated industries, and they're the ones that have been hardest to automate responsibly. Vertesia’s Process Engine applies agents where beneficial, and deterministic steps where needed.
The Model Context Protocol has become the industry standard for connecting AI agents to external tools and services. With this release external applications can now connect to your Vertesia project via MCP.
Build your Vertesia application in Claude Code or Codex (for example), and those coding agents can now create everything you'd normally do inside Vertesia Studio, directly from the coding agent without having to go into Vertesia Studio.
Vertesia’s remote MCP exposes the same skills and tools that are behind the Studio Assistant, but for those working outside of Vertesia’s AI Studio.
We've been making a deliberate argument for the past year: enterprise AI transformation isn't about any single feature. It's about building a platform that's reliable enough, flexible enough, and trustworthy enough to handle real work.
These updates are a better experience for every user who touches the platform. Agents that operate autonomously on your schedule. Agentic browsing that reaches the systems no API can. A workflow engine that makes structured, auditable AI automation possible for the organizations that need it most. And an integration that enables any coding agent to build on Vertesia.
Our platform is ready for what you need to build.