About Town
Town is an AI-powered assistant that connects to your tools to do real work on your behalf -- triaging your inbox, drafting in your voice, managing your schedule, and running workflows across everything you use. It knows you deeply and gets sharper over time.
About the Role
The core technical challenge at Town is: context from everywhere, action anywhere. The assistant pulls from your email, calendar, Slack, docs, and connected tools to build a deep understanding of who you are and what you need -- then executes reliably across all of them.
Town's AI assistant goes beyond a solid harness and decent tool integrations. It has memory that compounds over time, a suggestion engine that discovers what to automate before you think to ask, and network effects where your assistant gets better because your colleagues are using Town too. These are the systems you'll build.
You'll be a foundational hire, working across the full backend -- from LLM orchestration and workflow execution to the identity and knowledge layer that makes every interaction smarter than the last. Everyone has direct influence over architecture decisions. There's no platform team to hand things off to -- you own it end to end. The environment is greenfield, and the problems are unlike traditional backend engineering: non-deterministic, cost-sensitive LLM workloads; real-time personalization across a growing knowledge graph; agent-to-agent protocols that work across organizational boundaries.
Example projects include...
Building the orchestration layer for agentic workflows
Building multi-provider LLM infrastructure
Rethinking email, calendar, and contacts as AI-native experiences
Designing the identity and knowledge layer
Architecting agent networks
Building the trust and autonomy system
Defining what service ownership
You might thrive in this role if you...
Have built and scaled backend systems through a company's high-growth arc -- you were there during the inflection, not after. We want to hear all your your concrete product scaling stories.
Are comfortable defining technical direction in ambiguous, fast-moving environments -- not just executing within one
Think in systems -- distributed systems, data pipelines, APIs, reliability. You're drawn to infrastructure-level problems and the tradeoffs that come with them.
Have tuned your AI-native coding workflow that makes you dramatically faster and more effective.
Have experience with LLM infrastructure, agentic systems, or real-time AI workloads in production
Care about trajectory over tenure. We're less interested in years of experience than how fast you got here and what you built along the way.
Are excited about working in-person, five days a week. We believe it matters at this stage.
Location
5 days per week in person, in either SF or NYC
