AI Product Engineer, Distru

• 5+ years of full-stack development experience, ideally in B2B SaaS applications
• Hands-on experience building multi-agent harnesses in production — not just prototypes or personal projects
• Experience building MCPs (Model Context Protocol) that are actually used in production
• Strong grasp of production system concerns: reliability, observability, permissions, and safe workflow execution
• Experience across product and engineering — you've operated as both. You should be comfortable owning scoping and decision-making, not just implementation
• Proficiency in modern technologies such as React, TypeScript, Node, PostgreSQL, GraphQL, and LangGraph, with the ability to quickly pick up new tools and frameworks
• Startup agility: experience at an early-stage startup (under 20 engineers) where you've operated effectively in ambiguous, unstructured environments
• Low ego, team-first mentality — you measure success by what shipped and whether it worked, not by internal politics
• Clear, proactive communicator — you raise concerns and propose solutions before being asked

Desirable

• We're looking for an AI Product Engineer, working directly with our CTO, to architect and build an agentic harness — think of it as the Claude Cowork for vertical-specific SaaS
• This product will bridge the tools businesses already use to run their operations (ERP, CRM, spreadsheets, calendars, email, analytics, compliance systems) and fundamentally change how they interface with their own data and systems
• This is not a "bolt AI onto existing features" role. You'll be building a new, standalone agentic product from the ground up, which means you'll operate as both the engineer and the product owner: scoping the problem, making architecture calls, and shipping without a lot of structure around you
• You'll report directly to the CTO and work closely with product, engineering, and design as this fits into Distru's broader AI-native roadmap
• Distru is at an inflection point. We have deep market penetration in cannabis ERP, a commerce layer live and growing, and a roadmap that's increasingly AI-native
• This role owns one of the most ambitious pieces of that roadmap — building the connective tissue that lets an entire industry run its business through an agentic interface
• Architect and build a multi-agent system in production — one capable of safely operating across a business's core tools (ERP, CRM, spreadsheets, calendars, email, analytics, compliance systems)
• Build and maintain MCPs (Model Context Protocol integrations) that connect the harness to real business systems
• Own production system concerns end-to-end: reliability, observability, permissions, and safe workflow execution — this is a system that will take real actions on customers' data, and the bar for getting it right is high
• Tech Stack: LangGraph, Next.js, React, Prisma, Claude Code
• Since there's no existing playbook here, you'll define scope, make the call on tradeoffs, and drive execution with minimal oversight
• Design for the user, not just the architecture — a product-first mindset that prioritizes how this actually gets used over what's technically interesting to build
• Proactively identify and raise risks or open questions before they become blockers, and bring structured solutions, not just problems
• Keep pace with a fast-moving space — agent frameworks, harness toolkits, MCP standards — and bring in better approaches as they emerge rather than waiting to be told
• Work hands-on with tools like Claude Code as part of your own build process