Close moves 7 million minutes of call recordings, millions of sales emails, and a steady volume of SMS through our pipes every month. The Communications team owns that entire surface - and the job got trickier in the last year, because real-time calling is now a real-time AI problem. When a customer tells Chloe to make a call, everything that happens from that moment on - the live call, the conversation itself - is this team's.
We're looking for someone energized by both sides of the role: the distributed-systems challenges and the AI challenges.
One thing to know up front: we do move people between teams as the work shifts. Most engineers here end up on more than one team over their time at Close - this team is where you'll start, but over time you'll likely have the opportunity to work on many different projects.
This role is open at multiple levels - Software Engineer, Senior, and Staff. You don't need to pick one when you apply: we'll calibrate together during the process.
You are- A seasoned engineer. Python is our backbone, but perhaps you've worked across Go, Rust, or TypeScript. You pick the right tool for the workload rather than retreating to what you know.
- Deeply experienced with email and calendar platforms. IMAP, sync protocols, calendar APIs, and the thousand edge cases of oddly configured enterprise mail servers. This is the experience that makes someone successful on this team fastest.
- Fluent in real-time and async distributed systems. WebRTC, WebSockets, queues, streaming. You've debugged a production incident where latency budgets didn't hold.
- Opinionated about AI in real-time workflows. You've built or used live transcription, voice agents, or in-the-loop AI features. You have a view on what's ready for production versus what's still a demo.
- At home in the gnarly bits. Telephony providers, codecs, retries, eventual consistency. You're drawn to the systems where edge cases actually matter. Deep Twilio or similar telephony experience is a big plus (not a requirement - but a game changer off the bat).
- Working with AI in your day-to-day. You use AI tools in your own workflow to ship faster, write tighter code, and reason about unfamiliar parts of the codebase.
- A builder first. You'd rather get a sloppy v1 in front of fifty customers than spend three weeks on abstractions. You ship.
You will- Overhaul how Close syncs email and calendars. Our sync engine has run for a decade; we're making a serious, ground-up investment in its performance and reliability. Millions of sales emails and calendar events, and when it breaks, customers feel it within ten minutes. You'd be in the middle of one of the biggest infrastructure rebuilds at Close.
- Take voice agents further into the call. We own the live conversation layer for Chloe's calls, and there's a lot of work ahead expanding our voice agent capabilities and infrastructure - including custom models tuned for conversational edge cases like voicemail and automated phone-tree detection (turns out "wait for the beep" is a hard AI problem).
- Solve inbound. So far Chloe makes calls; next she answers them. An inbound conversation can go anywhere - which raises real questions about specialized agents, a receptionist layer, and seamless call routing between them. Largely unexplored territory.
- Build native scheduling. Core calendar scheduling capabilities inside Close, plus more powerful scheduling for our agents. Various projects landing over the next 6-12 months.
- Design human orchestration of an agentic workforce. Agents create work for humans too - things to review, approve, respond to. How do we surface what's happening inside the CRM so each person engages with what actually needs them? A genuinely open product-engineering problem.
- Push connect rates and answer rates. Phone number pooling, local presence, branded calling, reputation monitoring. Real revenue impact, real telephony plumbing.
- Move our streaming and event substrate forward. Kafka, Redis, pub/sub. The plumbing the rest of the product runs on.
Tech you'll touch: Python, Twilio, WebRTC, WebSockets, AssemblyAI, ElevenLabs, Kafka, Redis, PostgreSQL, MongoDB.
