Senior Software Engineer, AI Platform (Remote)

Weedmaps is a cannabis technology company seeking a Senior Software Engineer to build and operate its AI Platform. The role focuses on production AI agents, retrieval and knowledge systems, evaluation, cost and latency optimization, security, observability, and mentoring engineers while partnering with Developer Experience, Data, and product engineering teams.

Responsibilities:

• Design, build and operate production AI agents, including an autonomous coding agent that works from Jira and Slack, opens PRs, fixes its own CI failures and applies review feedback
• Build the retrieval and context layer behind our "AI brain": embeddings and semantic search, knowledge-graph modeling, and integration with our Snowflake semantic layer and internal docs, so agents pull the right context in one pass instead of several
• Extend our internal MCP server with shared skills, tools and processes that other teams can reuse, and help product teams bring agents into their own workflows
• Own evaluation: build offline test sets and online metrics that show whether a change to a prompt, model or retrieval step made an agent better or worse, and gate releases on them
• Drive cost and latency down: route work to the cheapest model that can do it, use caching and structured outputs, and track spend per agent and per task on AWS Bedrock
• Harden agents against prompt injection, memory poisoning and over-broad permissions: least-privilege tool access, auditable actions, human approval where the risk warrants it
• Instrument agents with tracing and logging so failures can be diagnosed and fixed quickly, and take part in on-call for the systems you own
• Evaluate new models and tools with a practical eye, adopting what holds up and passing on what doesn't
• Write clear design docs, review code, and mentor other engineers on building with LLMs

Requirements:

• Bachelor's degree or equivalent practical experience
• Minimum of 6 years of professional software development experience, including owning production services end to end
• At least 1 year of hands-on experience shipping LLM-powered features or agents to production (not only prototypes): tool/function calling, prompt design, structured outputs, and handling model failures gracefully
• Experience building retrieval systems: embeddings, vector or hybrid search, chunking and ranking, and measuring retrieval quality
• Strong proficiency in at least one of Python, TypeScript or Ruby, and comfort working across a polyglot codebase
• Experience with a major cloud platform (AWS preferred), CI/CD, and observability tooling
• A track record of measuring what you build: evals, A/B tests or clear before/after metrics
• AWS Bedrock, and experience managing model cost and latency at scale
• Model Context Protocol (MCP) servers or other agent tool frameworks
• Knowledge graphs or semantic layers (for example, Snowflake semantic views, dbt metrics or graph databases)
• Agents that act on developer workflows: GitHub Actions, Jira and Slack integrations, automated code review or code generation
• LLM security: prompt-injection defenses, sandboxing, permission models and red-teaming
• Entity matching or classification at large scale (millions of records) using ML combined with LLM reasoning
• Experience with Ruby on Rails, Elixir or Node.js services
• Experience at a marketplace, e-commerce or regulated-industry company