Full Stack Software Engineer - AI Applications

About the role

An agent orchestrator, not a typist. You don't want to be a faster keyboard. You want to be a manager of agents — handing off the heavy lifting, reviewing the output, and keeping your hands on the architecture. The IDE is a cockpit for orchestration, not a text editor. A productively lazy software engineer. Your dream workflow: describe the requirement, let the agent build it, verify, ship, next. You automate anything a human shouldn't be doing twice. Your biggest bottleneck should be deciding what to build — while AI executes the how. A fundamentals-first software engineer. Data structures, algorithms, distributed systems, networking — you understand the machine, not just the library that wraps it. When a framework breaks, you fix it. When AI gives you the wrong answer, you catch it. Orchestrating agents only works if you can tell good output from garbage. A first-principles thinker with vision. You break problems to their core, question the assumptions, and rebuild. A software engineer's job is to architect solutions, not wrestle with syntax. You don't copy an architecture because "that's how it's done" — you ask why and decide if there's a better way. A high agency software engineer. You don't wait for perfect specs or permission. You find a path, propose it, and move. Large organizations have walls; you figure out which ones to go through, around, or remove — and you do it constructively. A bias-for-action software engineer. Requirements will be messy and priorities will shift. You ship v1, learn, and iterate instead of living in design review.

What You'll Do

Orchestrate agents to build. Use AI as your default way of working — agents do the heavy lifting, you direct, review, and harden. You set the bar for how the team builds with AI. Build the AI use cases. Design and ship AI-powered applications and agents — RAG pipelines, LLM integrations, agentic workflows. Understand what's happening under the hood and make it work in production, at scale. Ship cloud-native systems on GCP. Design, deploy, operate. You own the architecture, the infrastructure-as-code, and the CI/CD pipeline. No throwing code over the wall. Connect systems that don't want to be connected. Build the integration layer across enterprise SaaS platforms. Expect messy APIs, legacy constraints, and creative problem-solving. Automate what shouldn't be manual. If a human repeats it, you write the code — or point an agent at it — to stop it. Make other engineers faster. Build the tools, agent workflows, and guardrails that remove friction. Developer productivity is a multiplier.

What You Bring

Python — deep. You write production systems, not just scripts. JavaScript / TypeScript — React, Node.js, or equivalent. A real frontend and a real API. Front-end strength is a big plus. AI engineering — LLMs, embeddings, vector stores, RAG. Comfortable building agents and evaluation pipelines. Bonus for fine-tuning or eval work. AI-assisted development — fluent with modern AI coding tools and agent workflows, and clear-eyed about where they help and where they don't. GCP — Cloud Run, Cloud Functions, GKE, BigQuery, Cloud Storage. Deployed and operated, not just tutorials. Databases — relational and NoSQL. You know when to use what, and why. APIs & microservices — designed and built RESTful services at scale. Git, IaC & CI/CD — Terraform, Cloud Build, or equivalent. Reproducible, version-controlled infrastructure and disciplined code review.

Requirements

BS in Computer Science, Electrical Engineering, or a related field. MS is a plus, not a substitute for shipping software. 2+ years building and deploying full-stack applications in production. 2+ years on cloud-native platforms (GCP preferred). Demonstrable experience building AI agents or applications — show us what you've built, not what you've read.

Why This Role

Ford is a 120-year-old company moving fast on AI, and that's the real challenge: building world-class AI applications inside a large enterprise with real constraints — security, compliance, legacy systems. The engineers who thrive here treat those constraints as problems to solve, not reasons to stop. If you want a place where everything is already figured out, this isn't it. If you're a 10x thinker who wants to build something that matters — and have the grit to push through the hard parts — let's talk.