At Composio, we are building infrastructure that allows agents to communicate with the tools you use for work including Github, Gmail, Notion, Salesforce, etc. We are a small team of engineers wrangling problems from context to search, that help us provide the most capable bridge between your agents and your tools.
We raised a $25M Series A from Lightspeed with some incredible angels like Guillermo Rauch (CEO of Vercel), Dharmesh Shah (CTO of Hubspot), Gokul Rajaram. Beginning of this year we 3x our ARR, our customer range from your friends in the YC batch to Wabi, Glean, Zoom and many more.
Building Agents
Write down the manual playbook as you run it, then work out what a person should never have to touch again
Build AI agents that watch inboxes, send the follow-ups, and keep the tracker current on their own
Turn repetitive outreach and tracking into workflows that keep running without you in the loop
Multiply yourself, so one person can keep hundreds of apps moving at once without anything slipping
Operations
Shepherd new app connectors from first request to live: securing OAuth approvals and API access, spinning up sandbox credentials and test accounts, and clearing whatever verification or paperwork stands in the way
Hold hundreds of apps in view at once, each parked at a different stage of approval
Chase approvals until they actually land, not until the first email goes unanswered
Make the demo videos and scope-mapping docs that move partners from maybe to yes
Own the status of every app, so engineering, product, and partners never have to ask where things stand
You already build with AI every day. It's part of how you work, not a box you'd tick out of curiosity
You stay with a problem until it's actually solved, sending the fifth follow-up long after most people quit at the first
You can keep fifty-plus threads in motion at once without letting one slip through
You work well without a map: you find the real problem before anyone hands it to you, then go fix it
Your first instinct is to automate a task rather than do it a second time
Automation, agents, workflows, and operations are the tools you use to hit a goal, not the job itself. We want someone who cares about achieving the goal, not a "whatever it takes" task executor
On the practical side:
An engineering background — BTech, CS, or similar — ideally with a few projects you've actually built and shipped
Writing clean enough to land in a partner's or customer's inbox without a second draft
Able to be in Bengaluru, in person, for the full 2 months (we are willing to extend the internship tenure to 6 months if more projects come in)
Final-year students and recent grads equally welcome, including if you're holding a full-time offer that starts later in the year
