Three roles.
One window.
01 · you
the builder
an idea
02 · the AI
builds + ships
03 · your customers
many tenants
clanker-screener
memecoin scoring · paid tier
whale-tracker
wallets > 100 ETH alerts
farcaster-bots
auto-posters per project
airdrop-portal
claim + allowlist checks
pump-watch
early-launch monitor
tg-trade-bot
in-chat trading commands
You
Your harness
Your tenants
You drive — at your Cloud Desktop. The AI codes — inside your VM. Your tenants pay — in USDC, on the same machine.
the hidden cost
It isn’t the LLM.
It’s the glue.
one bill.
one Cloud Desktop.
the agent writes the workflow as code. cron, websockets, queues — all native, all on the same machine.
The LLM bill you would pay anyone. The orchestration tax wrkr deletes.
the orchestration
wrkr is the substrate.
The agent wires the rest.
wrkr gives you a Linux workstation in the cloud — one dedicated Firecracker microVM per person, the same KVM-based isolation tech that runs AWS Lambda and Fargate. Cloud Desktop on top. Your AI coding harness of choice (OpenCode, Codex, Claude Code, Gemini CLI) running natively inside. Multi-unit deploy, multi-tenant routing, persistent state, slug-routed public URLs. The harness writes your integrations — auth, payments, email, billing, the lot. Bring your preferred providers, or use wrkr's built-in rails.
The agent integrates whichever you pick. wrkr doesn't lock you to a stack — it runs the substrate and writes the glue.
the moat
AI codes on real Linux.
Comment on the preview.
Watch the agent drive Chrome.
Six capabilities the workstation substrate makes possible — Cloud Desktop, harness-neutral, real Linux, persistent state, multi-tenant deploy, your keys end-to-end. Every one a native primitive, every one earning its keep.
AI codes on real Linux.
Open a project — get a Cloud Desktop on your hosted Linux VM. Your AI coding harness of choice runs natively (OpenCode by default; Codex, Claude Code, Gemini CLI all welcome). A real terminal, an ext4 disk that persists, the package manager you actually need. The agent reads, plans, edits, runs builds. Every project is a folder on the same VM — no new machine, no tab fork.
Comment on the live preview.
Click any element on the served app. Type 'make this bigger' or 'change to emerald'. The agent edits source. The preview rebuilds — in seconds, on your hosted machine.
Watch the agent drive Chrome.
VM Chrome with a persistent profile per tenant. Logged-in sessions survive workspace reopens. The agent navigates, fills forms, scrapes — and one click hands the keyboard to you for 2FA, OTPs, captchas. Hand back when you're done.
Multi-tenant in one declared release.
One operator. Many crypto-native subdomains. Declare the release once — wrkr deploys per-tenant, each on its own custom domain. pumpwatch.io, whalealert.app, clankerscreener.xyz — all live, all isolated, all USDC-settled.
Web · worker · cron in one ship.
Real Linux on a dedicated Firecracker microVM — sub-second resume, real isolation, the same tech AWS Lambda runs on. Declare web + private worker + scheduled cron + storage as one multi-unit release group. wrkr supervises every unit, projects the public URL, keeps the private workers VM-local.
Your VM. Your keys. Your tenants.
BYOM (bring your own model) is a first-class primitive — your LLM API keys never leave your VM. Pay for plans in USDC from any wallet — no KYC at the substrate layer, no email tied to your account. Each tenant's Chrome profile lives in your VM, not in a shared cookie jar. wrkr never trains on your code, never reads your repos. Operator state is yours, hardware-isolated, end to end.