Our customers are sales teams who live in Telegram. They don’t file tickets; they tell you in chat that something’s off, and they expect it fixed while the deal is still warm. You’d be the engineer who sits closest to them: watch how a team actually sells, set up their pipelines, automations and campaigns, build what’s missing, and ship it into the product so the next customer gets it out of the box.
This is a production engineering role, not solutions engineering. You write the code, you deploy it, you own it after launch.
You’d work in a small team alongside frontend, backend and DevOps engineers, with AI tooling doing the heavy lifting. We’re betting on the same thing Sam Altman is:
We’re going to see 10-person companies with billion-dollar valuations pretty soon.
Sam Altman
We’re building Cavyro that way: a handful of people, each owning a big surface, shipping every week.
What you’ll do
- Embed with customers, map how their sales process really works, and turn that into working Cavyro setups.
- Ship across the whole stack: the Rails API, the React app, the Telegram service and the AI gateway.
- Build integrations against the tools customers already run, however messy their APIs are.
- Debug production issues with the customer in the chat, then turn the fix into a product improvement.
- Feed what you learn in the field straight into the roadmap. You’ll hear it first.
The stack you’d work on
- Backend: Rails 8 JSON API on PostgreSQL. Multi-tenant, background jobs on SolidQueue, Redis, billing through Paddle.
- Frontend: React 19 with TypeScript and Vite. TanStack Query, shadcn/ui, Tailwind, real-time over ActionCable.
- Micro services: TypeScript and Fastify. A GramJS (MTProto) mirror of real Telegram accounts plus a grammY bot, with Prisma, Postgres and Redis.
- AI: a TypeScript agent gateway built on the OpenAI Agents SDK, exposing the whole CRM over MCP.
You might be a fit if
- You’ve built and run production systems, and you’ve been the one on the hook when they broke.
- You’re comfortable in both Rails and React, or deep in one and fast to pick up the other.
- You can sit with a non-technical founder, understand their sales process, and explain your solution in their words.
- You’d rather own a problem end to end than hand it off between teams.
Big pluses
- You’ve built on GramJS, Telethon or raw MTProto, or shipped serious Telegram bots.
- You’ve worked with the OpenAI API in production: built agents, MCP servers, eval loops, or driven token spend down without losing quality.
- You’ve done customer-facing engineering before: forward deployed, implementation, or technical consulting.
What we offer
- 4-day work week.
- Equity with standard 4-year vesting and a 1-year cliff.
- 20 paid days off.
- 12 sick days, no doctor’s note needed.
- 1 or 2 team onsites a year.
- Paid AI tools: the coding agents and models you want to work with.
How to apply
Email careers@cavyro.com. Skip the cover letter; send something you’ve built and a few lines on what you’d want to own at Cavyro. We reply to everyone within a week.