NeuronEdge Enclave is now part of Mindpool — the execution boundary in a reference architecture for AI your organization owns.

/who-its-for

Who it's for

NeuronEdge Enclave is infrastructure for teams that need hardware-level agent isolation and an audit record regulators can verify independently.

role: platform-engineer

Lead persona

The Regulated-Enterprise Platform Engineer

the pain
Engineering wants to ship agents. Security says no containers. Compliance says no managed sandboxes. The internal answer is to manually review every agent action — which makes agents useless.
what they need
A hardware-isolated execution environment they own, with audit evidence for auditors, that runs in their VPC. There is no such product they can buy today.

trigger → Failed security review of E2B/Modal; new internal AI policy mandating attested execution; SOC2/HIPAA audit gap; an incident where an agent took an action it shouldn't have.

role: compliance-buyer

Compliance-led buyer

The Mid-Market Compliance-Driven Buyer

the pain
Same problem as the lead persona, but no platform team big enough to self-host. They want a product that satisfies their regulator, not an infrastructure project.
what they need
A hosted confidential agent runtime with regional residency. The compliance posture without the infrastructure project.

trigger → Regulatory pressure (AI Act, HIPAA, regional data-residency) makes attested execution a requirement, not a nice-to-have.

role: framework-builder

Open-source adopter

The Agent Framework Builder

the pain
Needs an execution backend for their framework. Managed SaaS only or AGPL licenses are blockers.
what they need
A well-documented, Apache-licensed sandbox they can wire into LangChain, Mastra, or custom frameworks.

trigger → Apache-2.0 license + polished SDKs + first-class framework adapters (ne-langchain, ne-llamaindex). No paywall on basic operations.

role: saas-builder

High-volume adopter

The SaaS Builder Optimizing Unit Cost

the pain
Runs millions of short-lived agent tasks. Cares about predictable margins and density.
what they need
The runtime for cost reasons; the control plane for multi-host orchestration. Not a confidential buyer.

trigger → Density + performance story. ~2ms warm-pool create validates unit economics at scale.

role: security-researcher

Technical champion

The Security Researcher / Threat Modeler

the pain
Evaluates sandboxes for red-team labs, malware handling, CTF infrastructure.
what they need
A published threat model and attestation chain they can review independently, line by line.

trigger → Open threat model at /security + independently verifiable, signed audit event stream.

/who-its-for/not

Who it's NOT for

not a fit

Hobbyist on macOS expecting native Firecracker

macOS is dev-only via remote Linux host. NeuronEdge Enclave does not promise production Firecracker on macOS.

not a fit

Customer wanting a drop-in E2B API replacement

Self-hosted first. Managed cloud is a separate future offering.

not a fit

Buyer of a fully turnkey AI platform

NeuronEdge Enclave is infrastructure. A control-plane integration layer is expected above it.

Ship agents into a hardware-isolated, auditable execution boundary.