On-Site: Forward Deployed Engineer (FDE) _ Englewood Cliffs, NJ
On-Site in Englewood Cliffs, NJ
Experience: 5-10 years
Video Interview
12 Months Contract
The role
You are a software engineer, not a consultant. You embed directly with a customer — joining their standups, working with their data, inside their environment — and ship working AI solutions, not slides. Your motto: one customer, many capabilities.
You will use our enterprise AI platform to assemble solutions in the field (data pipelines, ontology, RAG-grounded agents, applications), or our AI-SDLC framework to run AI-driven software builds — and often both. The bar you'll be held to: a working application on real customer data within the first ~10 days of an engagement.
What you'll do
- Embed on-site or deeply with customer teams; map their most painful workflows with your Deployment Strategist partner and turn them into shipped software.
- Build data pipelines (e.g., PySpark) connecting customer source systems — ERP, CRM, databases, documents — with quality, lineage, and row/column-level permissions.
- Model customer data as an executable ontology (Objects · Properties · Links · Actions) and prepare it for AI consumption (vector indexes, retrieval sources).
- Build RAG-grounded agents and applications (e.g., TypeScript) that don't just answer — they act, writing back to real systems via tool-calling.
- Run fast validation loops with business users: weekly iterations, evaluation, guardrail tuning, go/no-go.
- Harden and deploy to production with the Core-Dev team — multi-cloud (AWS/Azure/GCP) or fully on-premises/air-gapped.
- Deliver AI-driven software builds (new applications, SaaS replacements, migrations) using our AI-development harness — orchestrating specialist AI agents through a gated lifecycle rather than hand-writing every line.
- Feed what you learn back to HQ: patterns you validate in the field become standard platform components.
What we're looking for
- Strong software engineering fundamentals — you can design, build, debug, and ship production systems end to end.
- Hands-on experience with LLM applications: RAG, agents, tool/function calling, prompt and context engineering, evaluation.
- Data engineering competence: pipelines, SQL, data modeling; PySpark or similar a plus.
- Full-stack ability to stand up usable applications quickly (TypeScript/React or similar).
- Comfort operating in ambiguity at a customer site — extracting requirements from real users, making scoping calls, and defending technical decisions to non-engineers.
- Bias for shipping: you'd rather demo something real in ten days than perfect something in three months.
- Excellent communication; you will be the face of the team at the customer.
Nice to have
- Kubernetes/Helm/Terraform familiarity; experience deploying in restricted or air-gapped environments.
- Experience with AI coding agents/harnesses (Claude Code or similar) used for production-grade development.
- Ontology, knowledge-graph, or enterprise data-platform experience.
- Enterprise domain exposure: supply chain, CRM, HR systems, e-commerce, or manufacturing.
- Korean language ability (many stakeholders are LG affiliates) — not required.