Forward Deployed Engineer, Enterprise AI Integrations
Wire LLM agents into the ERPs, CRMs, and ticketing systems enterprises already run, inside the client environment, and make the integrations survive production.
- Python
- TypeScript
- NetSuite
- Salesforce
- MCP
- AWS
Most enterprise AI projects die in the integration layer, not the model layer. The agent works in a demo and then meets NetSuite, Salesforce, SAP, Zendesk, and a dozen internal services with no documentation and three flavors of authentication. This role exists to get through that layer with the client, in their environment.
You embed with the client for the engagement, map what their systems actually expose versus what the vendor docs claim, and build the tool layer, the data contracts, and the failure handling that let an agent act inside those systems without corrupting a record.
Map the real system
- Run discovery inside the client environment: audit the APIs, event streams, and batch jobs their systems really expose, and write down the gaps before they become change orders
- Design the tool surface an agent gets over each system, with idempotent writes, explicit confirmation steps for anything irreversible, and audit trails the compliance team will accept
- Build the reconciliation checks that catch drift between what the agent believes and what the system of record says
Ship inside their walls
- Deploy into the client cloud accounts and identity boundary on AWS, GCP, or Azure, using their secrets management, network policy, and logging rather than bringing our own
- Handle the unglamorous parts: rate limits, OAuth token lifecycles, retry budgets, and the vendor sandbox that behaves nothing like production
- Stand up evals and regression runs against sanitized production data so a model upgrade cannot quietly break a workflow
Leave it operable
- Write the runbooks, dashboards, and alert routing the client on call team will use at 3am
- Train the client engineers on the tool layer until they can add the next integration without us
- Report honestly on what is automated, what is assisted, and what still needs a human
Required
- Six or more years building integration heavy backend systems in TypeScript, Python, or Go
- You have shipped at least one LLM or agent system into an enterprise production environment and lived with the consequences
- Real experience with enterprise platforms such as NetSuite, Salesforce, SAP, Workday, or Zendesk, including their API quirks
- Comfort working inside a client security, identity, and change management process rather than around it
- Direct written communication with stakeholders who do not share your context
Helpful, not required
- Experience with MCP servers, function calling schemas, or agent frameworks in production
- Background in professional services, solutions engineering, or a forward deployed engineering team
- Security or compliance exposure: SOC 2 controls, data residency, and PII handling in LLM pipelines
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, Agentic Commerce
Remote Applied AI
Embed with commerce clients and ship agent driven discovery, catalog, and checkout systems into production on their stack, not ours.
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