How it works Pricing Docs Blog Get Started →
Autonomous AI engineer · Runs 24/7 · Browser + Terminal

The AI engineer that ships
while you sleep.

Give Sweet a goal and walk away. It reads your repo, plans the work, writes the code, runs the tests — and messages you when it's done. Reply from your laptop or your phone, like texting a teammate who never clocks out.

Start shipping → $20/mo Watch it work ↓
🔒 Stripe checkout· Cancel anytime· No API keys needed· 2-minute setup
React· Next.js· Python· Node· Rails· Go· Rust· Any stack
Real session
Watch it work.
Sweet reasons through the problem, explores your codebase, writes the code, and runs the tests — all in one pass.
sweet — building auth system
› build a JWT auth system with login, register, and protected routes

read_file package.json, src/routes/index.ts, prisma/schema.prisma read_file src/middleware/*, src/models/user.ts
💭 Reasoning: The app uses Express + Prisma. No auth middleware exists yet. I'll add bcrypt + jsonwebtoken, create the User model migration, build the auth routes, and add a protect() middleware. Tests with Jest + supertest.
write_file src/auth/jwt.ts — JWT sign/verify helpers write_file src/auth/middleware.ts — protect() route guard write_file src/routes/auth.ts — POST /login, /register run_command npx prisma migrate dev --name add_users write_file tests/auth.test.ts — 12 test cases run_command npm test
✓ 12 tests passing
✓ 6 files written  →  auth/jwt.ts, auth/middleware.ts, routes/auth.ts, migrations/…
time: 4m 38s · tokens: 218k · cost: $1.41
Get your agent → $20/mo
Always on
Message it like a teammate.
It works around the clock.
Sweet doesn't stop when you close the tab. It keeps working through its queue, checks its inbox between tasks, and sleeps when there's nothing to do. Your messages work like email — send anytime, from any device.
Your agent — working thread #4
Ship the pricing page tonight, and add the annual toggle we talked about
11:42 PM✓✓
Done — pricing page deployed with the annual toggle (17% discount badge). Also fixed a layout shift on mobile while I was in there. Diff is on the pricing-v2 branch.
2:13 AM
legend. tomorrow let's do the onboarding emails
7:58 AM✓✓
End-to-end, not autocomplete
Features, tests, migrations, deploys — it finishes the job, not the sentence.
2-minute setup
Subscribe and run your first task in the browser. No config, no API keys.
$20/mo flat
Hard usage cap, no overage bills, cancel in one click.
How it works
Subscribe. Describe. Ship.
Works in your browser or your terminal. Start in minutes, cancel anytime.
Step 01

Describe the outcome

Tell Sweet what you want shipped. A feature, a refactor, a bug fix, a migration. Plain English works.

Step 02

Sweet works autonomously

It reads your repo, plans the work, edits files, runs commands, and iterates until the task is done — even after you close the tab.

Step 03

Review and ship

Come back to a clean diff, passing tests, and a reply in your thread summarizing what changed. Merge when ready.

What you can do
Works on real engineering tasks.
No IDE lock-in, no per-seat pricing, no hand-holding. Just results.
🔧

Build features end to end

Give Sweet a goal. It reads the existing codebase, writes the implementation, updates tests, and wires everything together.

› add Stripe subscription billing with webhook handling and invoice history page
🐛

Fix bugs with context

Point Sweet at a failing test or error log. It traces the issue through your codebase and writes the fix with a regression test.

› users are getting logged out randomly — investigate the session handling
🧹

Refactor safely

Break up a god class, migrate a database schema, or modernize a legacy module — with checkpoints and tests to verify correctness.

› refactor the payment module from callbacks to async/await, keep all tests green
🌙

Leave it running overnight

Queue up the backlog before bed. Sweet works through it task by task and replies to each thread as it finishes.

› clear the backlog: fix linting errors, update deps, add missing tests — message me per item
Start Pro → $20/mo
Why Sweet
Built for autonomous work.
Sweet doesn't just suggest code — it ships entire features while you're away.

🤖 Full Autopilot

Give Sweet a goal and walk away. It plans, edits, tests, and iterates until done — no approval chains, no hand-holding.

🕐 Runs 24/7

One continuous session with automatic memory compaction. The agent decides when to work and when to idle — it never needs a restart.

⚡ Parallel Agent Swarms

Dispatch subagents for independent tasks. Refactor the backend while another agent builds the frontend — coordinated from one session.

🛠️ Browser and Terminal

A full web app for anywhere access, and a CLI that runs in your own repo with your own tools. Same agent, same subscription.

💰 Flat, Capped Pricing

$20/mo Pro or $200/mo Max, hard-capped. No surprise overages, transparent usage tracking, cancel in one click.

🔒 Works With Your Stack

Any language, any framework. It runs your linters, your test suite, your CI — and adapts to your conventions, not the other way around.

Pricing
One engineer's coffee budget.
An entire engineer's output.
Flat monthly price, hard usage cap, cancel anytime. No contracts, no overages, no API keys to manage.
Most popular
$20/month
Pro — built for daily engineering work.
  • ~5M tokens every month
  • Web app + CLI, all models included
  • 24/7 agent with message threads
  • Parallel subagent swarms
Start Pro →
For heavy users
$200/month
Max — production workloads and swarms.
  • ~50M tokens — 10× the Pro budget
  • Everything in Pro
  • Built for teams shipping at scale
  • Priority support
Start Max →
🔒 Secure Stripe checkout · Cancel anytime in one click · Usage hard-capped, never billed for overages
A junior engineer costs $8,000+/mo. Sweet works nights and weekends for $20.
FAQ
Questions, answered.

How does the 24/7 agent actually work?

Sweet runs continuously: it works its queue, checks its message inbox between tasks, and idles when there's nothing to do. Your messages are email-like — it reads them at its next inbox check and replies in the thread. Read receipts (✓✓) show when it's seen them.

Do I need my own API keys?

No. Your subscription includes model access — no keys to manage, no separate model bills, no configuration.

What's a token budget in practice?

Pro's ~5M tokens covers regular daily agent work for an individual. Max's ~50M is built for production workloads and parallel swarms. Usage is hard-capped — you can never be surprise-billed.

Is my code secure?

Sweet runs in your browser session or your own terminal. We don't store or train on your repositories.

Can I cancel anytime?

Yes — one click from your billing page. No phone calls, no retention hoops. You keep everything the agent shipped.

Web app or CLI — which should I use?

Both are included. The web app needs zero setup and works from any device, including your phone. The CLI runs in your own repo with your own toolchain. Same agent, same threads, same subscription.

What if it gets stuck or goes off the rails?

You can watch every command and diff live in the logs view, interrupt at any time, and steer with a message. Autopilot has guardrails that pause it if it stops making progress.

Do you offer team or volume pricing?

Yes — contact us for custom plans if your team needs higher limits or multiple seats.

███████╗██╗    ██╗███████╗███████╗████████╗   ██╗
██╔════╝██║    ██║██╔════╝██╔════╝╚══██╔══╝   ██║
███████╗██║ █╗ ██║█████╗  █████╗     ██║      ██║
╚════██║██║███╗██║██╔══╝  ██╔══╝     ██║      ╚═╝
███████║╚███╔███╔╝███████╗███████╗   ██║      ██╗
╚══════╝ ╚══╝╚══╝ ╚══════╝╚══════╝   ╚═╝      ╚═╝

Your agent could be working
ten minutes from now.

Subscribe, send it a task, close the laptop. It'll message you when it's done.

Start shipping → $20/mo Read the Docs

Not ready yet? Get the Sweet onboarding guide

No spam, ever. Unsubscribe in one click.

Start Pro → $20/mo Compare Plans