Introducing Forge

Meet Forge: an AI-native workspace that enables engineers to build, test, and version event-driven agentic workflows to automate repetitive technical operations — no endless glue code required.

80% of technical operations budgets are wasted on toil. TechOps teams spend the majority of their time clearning through the same repetitive tickets. Day after day, engineers fight fires, patch systems, and chase down tribal knowledge buried in scattered documentation and runbooks. The promise of automation through IFTTT (Zapier, n8n, Moveworks, etc) tools has been around for years, yet ITOps and DevOps teams still find themselves stuck doing repetitive, manual work instead of solving higher-order problems.

The tools that were supposed to help—like n8n, Tines, Moveworks, Flowise—fall short. IFTTT-style builders force engineers to spend hours designing brittle decision trees that break at the first sign of change. Agent-building tools are equally mismatched: technical operations tasks often require reasoning over code, logs, and unstructured data in real time, something these platforms simply can't handle. The result? Automation that's either too rigid or too naive to be trusted in production.

At a37, we're changing that with Forge—the first AI-native technical operations automation platform. Forge gives ops teams (SRE, DevOps, ITOps, SecOps) the power to build, test, and version event-driven automations that range from fully deterministic workflows to adaptive, agentic systems.

Furthermore, one of the biggest barriers to automation today isn't just building workflows—it's knowing what to automate in the first place. That's why Forge includes built-in process mining. It continuously analyzes logs, workflows, and operational data to uncover inefficiencies and surface new automation opportunities. Instead of guessing where to start, teams can target the highest-impact pain points, accelerating adoption and maximizing ROI from day one.

Teams are already using Forge to offload work like infrastructure management, patching and upgrade cycles, cost optimization, compliance operations, and automatic remediation. Because workflows are standardized, auditable, and stored as simple YAML in a `.forge` folder, they can be versioned, reviewed, and shared across teams—cutting overhead, ensuring best practices, and accelerating resolution of critical issues.

The experience is designed to feel effortless. Forge's AI builder figures out what you're trying to accomplish, pulls in the context it needs, and helps you assemble workflows quickly—without endless glue code. Its runner can dynamically handle parameters for tool calls, retry failed steps, and even suggest fixes if something goes wrong.

Flexibility is built in. Some jobs call for step-by-step predictability; others benefit from adaptive reasoning, like an agent that can read logs, surface insights, or propose a fix. Forge supports both—and always with a safety net. Agents default to read-only. They can propose pull requests, flag issues, or suggest changes, but actions that affect production require explicit human approval. Control stays firmly in your hands.

Workflows can be triggered by events, APIs, Slack commands, or manual runs—so Forge adapts to the way your team already works. Every action is observable and traceable, with logs that explain what happened and why. After each run, workflows can suggest optimizations, helping teams continuously improve their processes with almost no effort.

Forge represents a shift from automation that barely keeps up to automation that drives operations forward. By enabling truly self-driving technical operations, Forge frees engineers to focus on meaningful work, while the repetitive and routine is handled automatically, reliably, and safely.

Ready to transform your DevOps workflows?
Join the Forge private beta today