Introducing Workflow Automator v2: Smarter Automation for Teams

A New Era of Workflow Automation

Today we are thrilled to announce the general availability of Workflow Automator v2, the most significant update to our automation platform since its initial launch eighteen months ago. This release represents months of intensive development, driven by feedback from over two thousand customers who rely on Workflow Automator to streamline their operations every day. We have fundamentally rethought how teams build, test, and deploy automated workflows, and the result is a product that is faster, smarter, and more intuitive than ever before.

When we launched the original Workflow Automator, our goal was to make it possible for any team—not just engineers—to automate repetitive processes. That vision has not changed, but the way we deliver on it has evolved dramatically. Workflow Automator v2 introduces three major capabilities that our customers have been requesting: AI-powered workflow suggestions that help users build automations faster, a parallel execution engine that dramatically reduces workflow run times, and a completely redesigned visual builder that makes complex workflows easier to understand and maintain.

Before diving into the details of each new feature, I want to share some context on the scale of automation that our customers have achieved with the platform. Since launch, Workflow Automator has executed over fifty million workflow runs, saving our customers an estimated three million hours of manual work. The average customer has built twenty-seven active workflows, with power users maintaining libraries of over two hundred. These numbers underscore the critical role that automation plays in modern business operations, and they motivate our continued investment in making the platform more powerful and accessible.

AI-Powered Workflow Suggestions

One of the most common pieces of feedback we received from new users was that getting started with automation felt overwhelming. Even with our template library and documentation, users often struggled to identify which processes to automate first and how to structure their workflows optimally. Workflow Automator v2 addresses this challenge head-on with an AI-powered suggestion engine that analyzes your team's existing tools, processes, and data patterns to recommend high-impact automation opportunities.

The suggestion engine works by integrating with your existing tools—project management systems, communication platforms, code repositories, and more—and observing patterns in how your team works. It identifies repetitive sequences of actions, bottlenecks where work frequently stalls, and manual handoffs between team members that could be automated. Based on this analysis, it generates specific workflow recommendations complete with pre-configured steps, trigger conditions, and suggested integrations. Users can review these suggestions, customize them to their needs, and deploy them with just a few clicks.

The AI engine uses a combination of process mining techniques and large language model capabilities to generate its recommendations. Here is an example of how you can programmatically interact with the suggestion API:

import { WorkflowAutomator } from '@primates/automator-sdk';

const automator = new WorkflowAutomator({
  apiKey: process.env.PRIMATES_API_KEY,
  workspace: 'my-workspace'
});

// Fetch AI-generated workflow suggestions
const suggestions = await automator.suggestions.list({
  category: 'engineering',
  minConfidence: 0.85,
  limit: 10
});

for (const suggestion of suggestions) {
  console.log(`Suggestion: ${suggestion.title}`);
  console.log(`Estimated time saved: ${suggestion.estimatedTimeSaved} hrs/week`);
  console.log(`Confidence: ${(suggestion.confidence * 100).toFixed(1)}%`);
  console.log(`Steps: ${suggestion.steps.length}`);
  console.log('---');
}

// Deploy a suggestion as a new workflow
const workflow = await automator.workflows.createFromSuggestion({
  suggestionId: suggestions[0].id,
  name: 'Auto-deploy on PR merge',
  enabled: true
});

Parallel Execution Engine

The second major feature in v2 is our new parallel execution engine. In v1, workflow steps executed sequentially by default, which meant that workflows with independent steps—such as sending notifications to multiple channels or performing checks across several systems—ran much slower than necessary. The new parallel execution engine automatically identifies steps that have no data dependencies and runs them concurrently, dramatically reducing end-to-end workflow execution time.

In our internal benchmarks, enabling parallel execution reduced average workflow run time by sixty-two percent for workflows with five or more steps. For complex workflows with twenty or more steps, the improvement was even more dramatic, with some workflows completing in one-fifth of their previous run time. The parallel execution engine is fully backward compatible—existing workflows will automatically benefit from parallel execution without any changes required from users.

The execution engine implements a dependency graph model where each step declares its inputs and outputs. The engine analyzes these declarations to build a directed acyclic graph of step dependencies, then schedules steps for execution as soon as all their dependencies have been satisfied. This approach gives users the simplicity of defining steps in a linear sequence while automatically extracting parallelism wherever possible. Key features of the new engine include:

• Automatic dependency detection that analyzes data flow between steps and identifies which steps can run in parallel without any manual configuration required.
• Configurable concurrency limits that allow users to control the maximum number of steps executing simultaneously, which is important for workflows that interact with rate-limited external APIs.
• Graceful error handling with configurable retry policies, timeout settings, and fallback actions that ensure workflow reliability even when individual steps fail.

"Workflow Automator v2 has cut our deployment pipeline time from forty-five minutes to under twelve minutes. The parallel execution engine alone justified the upgrade for our team." — Chris Martinez, VP of Engineering at DataFlow Inc.

Redesigned Visual Builder

The third pillar of the v2 release is a completely redesigned visual workflow builder. We rebuilt the builder from the ground up using a modern canvas-based rendering engine that supports smooth zooming, panning, and drag-and-drop interactions. The new builder also introduces several features that make it easier to work with complex workflows, including collapsible sub-workflows, inline step previews, and a real-time collaboration mode that allows multiple team members to edit the same workflow simultaneously.

The visual builder now supports a split-pane view where users can see the visual workflow graph alongside the generated configuration code. This is particularly valuable for technical users who want to understand exactly what the visual builder is producing, and for teams that manage their workflow configurations in version control. Changes made in either pane are instantly reflected in the other, providing a seamless experience regardless of whether users prefer visual or code-based editing.

We have also significantly expanded our integration library. Workflow Automator v2 ships with over one hundred and fifty pre-built integrations spanning categories including cloud infrastructure, developer tools, communication platforms, project management, databases, and monitoring systems. Each integration comes with a set of pre-configured actions and triggers that can be dropped into workflows with minimal configuration. The following table highlights some of the new integrations available in v2:

Migration Guide and Pricing

Upgrading from Workflow Automator v1 to v2 is straightforward. We have built an automated migration tool that analyzes your existing workflows, converts them to the v2 format, and validates that they will behave identically after migration. The migration tool handles the conversion of step configurations, trigger definitions, error handling policies, and integration credentials. We recommend running the migration tool in dry-run mode first to review the proposed changes before applying them to your production workflows.

For our existing customers, Workflow Automator v2 is available at no additional cost on all current plans. The AI suggestion feature requires the Pro plan or above, while parallel execution and the redesigned builder are available on all plans including Free. New customers can try the full feature set with a fourteen-day free trial that includes all Pro plan capabilities. For teams interested in exploring what v2 can do, we have published a comprehensive migration guide and a collection of sample workflows in our documentation. Please note that the legacy webhook builder from v1 will be deprecated in six months and will no longer be supported after that period.

We are incredibly excited about the possibilities that Workflow Automator v2 opens up for our customers. The combination of AI-powered suggestions, parallel execution, and the redesigned builder represents a significant step forward in making automation accessible to every team. We have a packed roadmap for the rest of the year, with features like conditional branching, custom function support, and enterprise audit logging all in active development. Stay tuned for more updates, and as always, we would love to hear your feedback.

1. Sign up for a free trial at primates.dev/automator to explore the new features.
2. Review the migration guide if you are upgrading from v1.
3. Join our community Discord to share feedback and connect with other users.
4. Attend our upcoming webinar on March 15 for a live demo of v2 capabilities.