Twill
Twill is an AI-native project management platform that connects directly to a team's codebase and workflow tools, enabling software teams to delegate development tasks to AI coding agents. The platform supports mid-task handoffs between local and cloud environments without requiring commits or re-prompting, preserving branch state and conversation context. Teams can trigger agents from Slack, Linear, or GitHub, with cloud plans priced per workspace rather than per seat.
- Developer Tools
- AI Agents
- Artificial Intelligence
- Software Only
- Enterprise Software
Funding
Founders
Founder details are not available yet.
Product
Problem
Software engineering teams often struggle to translate high-level intent into actionable specifications and tickets, leading to miscommunication and slow execution. Managing the full development loop—from planning to code review—across multiple tools like GitHub, Slack, and Linear creates friction and delays.
Solution
Twill is an AI-native project management platform that connects directly to a team's codebase and workflow tools. It clarifies engineering intent into detailed specifications, creates and manages project tickets, and dispatches work to AI coding agents that execute in sandboxed environments. Twill supports seamless mid-task handoffs between local and cloud environments without requiring commits or re-prompting, preserving branch state, uncommitted changes, and conversation context. Agents can run scheduled tasks via cron-based triggers, open pull requests on separate branches for human review, and post updates to Slack, Linear, or GitHub where teams already work. The platform integrates with GitHub for code access, Slack for communication, Linear and Notion for issue tracking, and Sentry for error monitoring. Teams can run agents locally for free with their own AI subscriptions, or use Twill's cloud infrastructure priced per workspace.
Target Audience
Primary users are software engineering teams and engineering managers who want to automate development workflows, clear technical debt, and improve their AI coding agents' performance.
Features
- Runs any AI model, allowing teams to choose providers or bring their own API keys (BYOK) for Anthropic, OpenAI, or Google
- Supports multi-repository setups with full infrastructure and dev environment configuration, giving every task an isolated copy with files, processes, and services
- Connects to GitHub, Slack, Linear, Notion, Sentry, AWS, and GCP for a unified workflow
- Enables mid-task handoff between local and cloud environments without commit, push, or re-prompting, preserving branch state, uncommitted changes, and conversation context
- Supports scheduled and automated agent tasks via cron-based triggers, with agents opening pull requests and posting updates where teams already work
- Provides a desktop app (macOS, Windows, Linux), web interface, and CLI for flexible access
- Encrypts BYOK API keys and OAuth integration tokens at rest using authenticated encryption