
Jervis Boost
AI engineering control plane
Gives every AI coding agent on a team the same skills, roles, quality gates and project memory — from one CLI, an MCP server and a local dashboard.
Independent work
Jervis is where I build the tools I want to exist — an independent ecosystem of developer tooling and AI engineering infrastructure. Each layer is useful on its own and is designed to plug into the next.

AI engineering control plane
Gives every AI coding agent on a team the same skills, roles, quality gates and project memory — from one CLI, an MCP server and a local dashboard.

Secure execution layer for AI assistants
Lets AI assistants work inside a developer's real projects — files, git, processes, browser — over the Model Context Protocol, with read-only defaults, per-workspace permissions and an audit trail.
Local-first agent runtime
A provider-agnostic assistant and coding-agent runtime: routing and delegation across Thinker, Executor, Reviewer and Verifier roles, isolated git worktrees and persistent memory.
Self-hosted deployment platform
Deploy and operate apps on your own servers from one dashboard: a Go control plane, lightweight agents that only dial out, and a bilingual Nuxt 4 interface.
Arabic-aware math input
Math input and computation with first-class Arabic variables, shipped as one package for Vue, React, Angular and Web Components.
UI and runtime packages shared across Jervis products and published on npm.