# Member of Technical Staff - Honcho Core Engineer Core Engineer for Honcho, continual learning infrastructure for AI memory and identity modeling. Own the core, the ecosystem surface, and the machinery that tells us what's working. NYC, IRL. Date: 2026-08-12 URL: . Tags: positions, engineering, announcements --- NYC, Full-Time, In-Person The Situation Honcho is ripping. 50x developer growth in three months. 40,000 prosumers and developers. 600+ startups applied to our startup program in eight weeks. We're a product-coupled neolab solving personal identity for the agentic world. The market currently calls what we do memory, but the vision for Honcho is much more ambitious. We build better models of you than you have of yourself. The goal is true 1:1 individual alignment for every human extending their cognition and agency with artificial intelligence. That vision has an infrastructure bill attached. A model of every individual user, updated continuously, queryable in the loop of someone else's agent, is a genuinely hard systems problem--stateful, write-heavy, and unforgiving at scale. Growth like ours makes every architectural decision load-bearing, and we're making those decisions now. The repo has its own gravity, too. Honcho is open source, and we maintain integrations with Claude Code, Hermes, OpenClaw, Codex, and more arriving faster than we can staff them. The agent ecosystem is moving quickly enough that this is a permanent condition, not a backlog we clear. We need more talented people who can hold the whole thing in their head. The Job You'll work directly with the CTO to architect and maintain Honcho: continual learning infrastructure for AI memory and identity modeling. Own the core. The systems that store, update, and serve per-user models are yours. Not a service you contribute to, but the thing itself, from high-level architecture down to the implementation. Design the surface developers live on. Honcho is a developer product. Every API decision is felt immediately by 40,000 people building against it, and ergonomics compound in both directions. You'll be judged on the interfaces as much as the internals. Own where Honcho meets the ecosystem. Claude Code, Hermes, OpenClaw, and whatever ships next month--every one an upstream we don't control, moving on someone else's release schedule. You'll keep Honcho current across all of them without letting the integration layer fragment into a dozen special cases, and you'll do it in public: reviewing contributions, arguing architecture in the open, treating the community as an extension of the team rather than a queue to answer. Make the product measurable. Identity modeling is an empirical problem, and you can't improve what you can't observe. Working alongside research, you'll build the instrumentation and experiment infrastructure that lets us ship a variant of the memory pipeline and know whether it was better. Research shapes the questions; you build the machinery that answers them, and you understand the questions well enough to argue about them. Product and backend depth is the center of this role, but we need someone versatile enough to fill gaps wherever they open. You'll constantly trade performance against new features against developer experience, rarely get all three, and have nobody to arbitrate for you. You make the call, ship it, and own what happens next. That's a wide surface, and we know it. We don't expect equal strength across all four: tell us which one you'd own on day one and which you'd grow into. You Deep systems intuition. You've designed distributed systems that survived contact with real traffic, and you can reason about how they fail before they do. You think at multiple altitudes. Global and local, architecture and implementation, and you can synthesize between them without losing either. You've owned things end-to-end. Design, ship, deploy, operate, optimize, iterate. A track record of projects you carried the whole way. You've worked in public. Maintained something people depended on, reviewed contributions from strangers, and held a technical line in a thread without making enemies. You're empirical. You instrument before you optimize, and you're suspicious of improvements you can't measure. You work well with broad freedom. High agency, significant responsibility, minimal supervision. That's the setup, and it suits you. You understand the developer mind. You can translate user needs into product decisions and know what makes an SDK a pleasure or a chore. High conviction. This is a mission-driven team and a long haul. We want you in for it. NYC, in person--or ready to move. Technical Requirements 5+ years in similar roles, or equivalent Python backend development, with FastAPI or similar Ergonomic API design--you've built interfaces other developers had to live with SQL databases and data modeling Deep experience with distributed systems design and implementation Distributed state, caching, and data replication Containerization and orchestration: Kubernetes, Nomad, or equivalent Building and maintaining microservice architectures High-throughput, large-scale network systems Building with and integrating LLM-powered features Harness engineering. You've worked at the layer where agents actually run (context assembly, tool wiring, session state) and you understand it well enough to integrate into someone else's runtime without fighting it Evals and experiment design: you can instrument a system so a result means something Enough front-end to turn a mockup into a shipped feature A plus: cryptography, machine learning, or identity systems. Distributed consensus protocols and eventual consistency. Interests in the cognitive sciences--linguistics, neuroscience, philosophy, psychology. Up-to-date in the open-source AI community. Compensation Base: $180K–$215K + equity Plus: standard Plastic benefits Apply Two things: A system you designed that had to hold up under real load. Tell us what you got right and what you'd rebuild. Your record in whatever form it exists--resume, portfolio, GitHub, personal site, etc. APPLY HERE (Back to Working at Plastic)