Company · WorkingAgents — Who is building the AI Agent Gateway

§ 1.0

What the company is.

Three facts about what this company is and is not. Each statement is structural, not aspirational.

Categoryenterprise

An enterprise infrastructure company. Not a product company, not a services company — a company built to ship and support control-layer infrastructure that customers run inside their own environments.

Shapelayer

A control layer, not an agent builder. AI Agent Gateway does not build the agents. It authorizes the ones already running in production -- on any framework, on any model, against any downstream system.

Marketcategory

Execution control and security infrastructure for agentic systems — a new category adjacent to IAM, DLP, and gateway categories, distinct from each. Every comparison ends at the same place: we operate at a different layer.

§ 2.0

Why now.

The category existed in partial form for decades. Agentic systems made a dedicated control layer necessary.

Agents take action

Previous generations of AI produced text. Agents produce actions against enterprise systems — API calls, database writes, tool invocations. The action is the unit that must be enforced.

Existing controls sit below the action

IAM governs identities. DLP governs data. SIEM records events. None of them enforce autonomous agent actions at the moment they happen. The action layer was never named, because it was never populated — until now.

Procurement requires proof

Agentic systems reach production under the security review. Control becomes the gate. The company exists to be the layer a security team can point to before it signs the deployment off.

§ 3.0

Team.

The company is building under small-team discipline. Two founders cover engineering and operations.

Founder · Engineering

James Aspinwall

Engineer focused on infrastructure, distributed systems, and secure execution environments. Built systems where reliability, access control, and system integrity are enforced, not assumed. Focus: designing the execution layer that evaluates and enforces every agent action.

Founder · Operations

Liem Chatjellis

Operator with Director-level background in supply chain and enterprise systems. Built and deployed execution frameworks in environments where control, accountability, and process determine outcomes. Focus: translating real-world operational constraints into enforceable agent execution.

§ 4.0

Company principles.

Four principles govern how the company operates. They hold across product, security posture, and company communication.

01 Control before scale. Enforcement ships before the customer count grows; growth follows a verified control surface.

02 Trust through enforcement. Every claim maps to a mechanism; unverifiable posture language is omitted.

03 Document discipline. Revision strips, section numbers, and release SHAs reflect real revisions — document cues are earned, not decorative.

04 Infrastructure over theatrics. Depth and restraint beat polish and motion — the work is the product, the site reflects the work.

§ 5.0 — Continue

Three conversations. One company.

Whichever role brings this question, the next step is a direct conversation with the right counterpart.

For the CEO / COO / investor

Talk to us

A direct conversation about category, deployment scope, operating implications, and where the company is in its build. No SDR, no qualification call.

Open the conversation →

For the VP Platform / sponsor

Read the executive guide

Why agents created a new control problem, what the AI Agent Gateway is, and how to bring it through procurement. Written for the buyer.

Open the guide →

Who is building the AI Agent Gateway.

What the company is.

Why now.

Team.

Company principles.

Three conversations. One company.