Forward Deployed Engineer, Agentic Commerce
Embed with commerce clients and ship agent driven discovery, catalog, and checkout systems into production on their stack, not ours.
- TypeScript
- Python
- Shopify
- MCP
- Claude API
- Postgres
Forward deployed means you sit inside the client engineering org for the length of the engagement. You work in their repos, their cloud accounts, and their release process, and you own getting agentic commerce features from a scoped SOW to running in production with their team able to operate it after you leave.
The work is concrete: agent driven product discovery over Shopify Storefront and MCP endpoints, catalog and pricing tools a model can call safely, and checkout flows where an agent acts on behalf of a customer without ever being able to do the wrong thing. You will spend as much time on guardrails and evals as on the happy path.
Embed and ship
- Join the client team for the engagement, work in their codebase and deployment pipeline, and ship agent features behind their existing review and release gates
- Build tool surfaces over the Shopify Storefront API, Hydrogen, and headless catalog and inventory services that a model can call with typed inputs and bounded side effects
- Own the last mile: feature flags, staged rollout, and the rollback plan the client on call team can execute without you
Make agents safe to run in a store
- Design the permission model for what an agent may read, quote, reserve, and purchase, and enforce it in the service layer rather than in the prompt
- Write evals against real catalog data and real customer intents, and track regressions on every model or prompt change
- Instrument every tool call so the client can answer why the agent did something from logs, not from guesswork
Transfer ownership
- Document the system so the client engineers can extend it, and pair with them until they do
- Run the handoff: runbooks, alert thresholds, and a clear list of what is safe to change and what is not
- Feed what broke in the field back into our reference architecture
Required
- Five or more years shipping production software in TypeScript or Python, with at least one system built on LLM tool calling that real users depended on
- You have worked embedded in a client or partner engineering team and are comfortable being the outsider who has to earn commit access
- Commerce platform depth on Shopify Plus or a headless stack: Storefront API, Hydrogen, commercetools, or Salesforce Commerce Cloud
- You treat evals and observability as part of the feature, not a follow up ticket
- You can explain a model failure mode to a merchandising lead and a rate limit to a CTO in the same meeting
Helpful, not required
- Experience with the Model Context Protocol or building tool servers that multiple agents consume
- Exposure to Shopify Storefront MCP, agentic checkout, or ChatGPT and Perplexity commerce surfaces
- Prior forward deployed, solutions engineering, or professional services background
Also hiring.
Engineering Lead and Client Partner
Remote Delivery
A player coach seat that combines technical lead, delivery owner, and second seat on sales.
Forward Deployed Engineer, Enterprise AI Integrations
Remote Applied AI
Wire LLM agents into the ERPs, CRMs, and ticketing systems enterprises already run, inside the client environment, and make the integrations survive production.
Forward Deployed Engineer, Private LLM Inference
Remote Applied AI
Stand up self hosted and private LLM inference inside client infrastructure, from model selection and serving stack to the agents that run on it, and hand it over running.
Show us something you built.
Send a resume or profile with a short note about work you have shipped. Links to code, systems, or launched projects tell us more than a cover letter ever will.
careers@contracollective.com