Full Stack Engineer

About the role

Mint Cannabis runs 39+ stores across six states. Our entire tech stack — the storefront, our ERP that runs the business, the inventory pipeline, the POS integration, and the K8s cluster on GCP that hosts the supporting services — is currently built and maintained by one developer. You'd be the second. That means real ownership from day one, an unusually wide view of how a vertically-integrated retail business actually works, and direct mentorship from someone who's already touched every layer of the stack.

What you'll work on

• Day-one focus:
  • Storefront features — Astro 5 + Tailwind 4, deployed to Cloudflare Pages. New pages, components, and UX flows on the customer-facing e-commerce site
  • ERP backend work — Python modules for the multi-company ERP, custom website routes, POS integration, internal admin tooling
  • Inventory & discount pipeline — Node service syncing product and inventory data into Redis + Postgres. Features that affect what customers see and pay
  • Deploys across Cloudflare Pages, Railway, and our GKE cluster on GCP — you'll ship to all three from week one
  • Bug-fixing and incident response across the stack — you'll learn the system by fixing it
  • Stretch areas (over 12 months): deeper platform ownership (GKE administration, CI/CD pipeline design, networking); LLM-powered features (we use agentic workflows heavily, in product and in our dev workflow); vendor integrations (Dutchie, PostHog, Google Business Profile, Telnyx SMS)

What we're looking for

• Required:
  • 5+ years building production web applications
  • Strong with TypeScript and comfortable in Python (or strong in Python with TS fluency)
  • Shipped to a modern SSR framework (Astro, Next, Remix, SvelteKit, or similar)
  • Comfortable with relational databases (Postgres preferred) and at least one cache layer (Redis, Memcached)
  • Integrated with at least one third-party API that has real consequences when it fails
  • Production GCP experience — gcloud CLI, IAM, service accounts, Cloud Storage, Cloud SQL or equivalent
  • Comfortable in Kubernetes (GKE preferred) — reads manifests, debugs pods with kubectl, understands services and ingress
  • Networking/IP/DNS fundamentals — TCP/IP, CIDR, routing, firewall rules; DNS records (A, AAAA, CNAME, MX, TXT, NS) and SPF/DKIM/DMARC; reads traceroute and dig +trace
  • Web hosting & edge infra — hands-on with 2+ of: Cloudflare (Pages/Workers/DNS/WAF), VPS/cPanel, AWS (Route 53, S3, CloudFront), GCP (Cloud DNS, Cloud Run, Load Balancing), Vercel/Netlify
  • Docker fluency — writes/debugs Dockerfiles, layer caching, multi-stage builds, container networking
  • YAML fluency — K8s manifests, GitHub Actions, Docker Compose, Helm values
  • Strong git workflow — branching, rebase vs. merge, conflict resolution, code review
  • CI/CD experience — built/maintained pipelines (GitHub Actions, GitLab CI, CircleCI, or similar)
  • Observability with Grafana + Loki (or equivalent) — dashboards, LogQL/PromQL, alerting, real incident diagnosis
  • Multi-cloud literacy — worked across 2+ providers (we run GCP + Cloudflare + Railway)
  • Solid SDLC knowledge end-to-end
  • Test-driven development as a habit — Playwright for storefront, pytest for Odoo modules
  • On-call triage discipline — owned production incidents end-to-end, can write the postmortem
  • Standard app-sec competencies — OWASP (XSS, SQLi, CSRF, auth pitfalls, secret hygiene)
  • LLM-specific security awareness — treats model output as untrusted, understands prompt injection, guardrails tool calls
  • Communication, documentation, presentation — explains decisions at the right altitude for engineers, ops, and execs
  • Articulates engineering reasoning — trade-offs, alternatives, the data behind the call
  • Real Big-O literacy — N+1 queries, denormalization trade-offs, when a hash map beats a tree
  • Reads logs and stack traces fluently; debugs by evidence, not guessing
  • Container orchestration depth — Deployments, StatefulSets, DaemonSets, Jobs, CronJobs, rolling updates, probes, resource limits

Bonus points

• Odoo experience at the developer level
• Cannabis industry experience
• Built/maintained a Dutchie, Metrc, or similar retail-system integration
• Owned a GKE cluster, written Helm charts, or set up complex CI/CD into Kubernetes
• Shipped a production feature backed by an LLM (prompt engineering, context management, MCP servers, evals)
• Tailwind, Storybook, Playwright
• Cloudflare Workers, Pages, or KVPrometheus, Tempo, or Wazuh SIEMPublished technical writing, talks, or open-source maintainership

How you work

• Debug by reading code and logs, not by guessing
• Verify your changes work before saying they do
• Ask questions instead of guessing when the requirement is unclear
• Comfortable being the only person on a problem, equally comfortable handing it off
• Prefers boring code that doesn't break to clever code that does

How we work

• Small team — you, the CTO, and the rest of the Mint org. Real ownership from week one
• Ship multiple times a day. Code review is the default, not a ceremony
• Light on-call rotation, shared with the CTO. Fewer than 2 pages per rotation