Company Description
The Role
You'll be the technical lead across our product portfolio — owning architecture, writing production code, and setting the engineering standard. This is a builder's seat, not a manager's. You'll work directly with the founder, ship into production weekly, and help define how AI agents are woven into every product we make.
What You'll Do
- Architect and build full-stack systems across multiple products (Next.js, Node.js, Python, PostgreSQL, vector stores)
- Design multi-agent AI pipelines using Claude, LangChain, and local inference (we run our own GPU infrastructure)
- Own production deployments end-to-end — Cloudflare, Vercel, hosting, monitoring, SEO
- Make hard architectural calls and document them (no guessing, no "we'll figure it out later")
- Lead future engineering hires and set the bar for code quality
- Move between codebases without losing context — accounting one day, cannabis the next, scouting the day after
What You Bring
- Proven ability to build and ship production web applications end-to-end (frontend, backend, infra)
- Deep comfort with modern React/Next.js, Node, and Python
- Real experience integrating LLMs into products — not just calling an API, but building agentic systems, RAG pipelines, memory architectures
- Strong opinions about clean architecture and the discipline to keep things simple
- Self-direction. You can take a one-line problem statement and come back with a plan
- Comfort with ambiguity and multiple concurrent projects
We don't care where you went to school or how many years you've been doing this. Show us what you've built.
Bonus Points
- Worked with multi-agent frameworks (LangGraph, CrewAI, custom)
- Vector DB experience (Qdrant, Pinecone, LanceDB)
- Local model hosting (llama.cpp, vLLM, Ollama)
- Built or shipped on Claude Code, Cursor, or similar AI-augmented dev workflows
- Background in any of: accounting/fintech, cannabis tech, sports analytics, hospitality software
Compensation: Prove It, Earn It, Own It
We don't believe in inflated salaries for unproven contributions. Our structure has three stages:
- Prove It — Paid trial period with clear deliverables. We both decide if it's a fit.
- Earn It — Competitive base + performance milestones tied to shipped work.