AI Engineer

Location: Remote (North American business hours) Type: Full-time

About Our Client

Our client is a software studio that designs, builds, and ships production applications for clients across manufacturing, industrial, and B2B SaaS. They take projects from scoping and estimation through architecture, delivery, and long-term support. Increasingly, the products they build have AI at their core: retrieval systems, document understanding, and LLM-powered features that have to be accurate, fast, and cost-aware in production. They are a small, senior team that values clear thinking, strong ownership, and shipping work they are proud of.

The Role

We are looking for an AI Engineer to own the AI layer of products end to end. This is not a research role and it is not a prompt-tinkering role. It is about taking AI capabilities (retrieval, embeddings, LLM reasoning, agents) and turning them into reliable features inside real applications that customers depend on.

You will work across SaaS products and client engagements, building things like RAG pipelines, hybrid search, document ingestion, and LLM-driven workflows. You will care as much about retrieval quality, latency, and token cost as you do about getting a demo to work. You should be comfortable owning the full path from architecture to a shipped, monitored feature.

What You Will Do

• Design and build retrieval-augmented generation (RAG) systems: ingestion, chunking, embedding, storage, retrieval, and reranking.
• Build and tune hybrid search combining vector similarity (pgvector) with keyword and metadata filtering.
• Integrate LLM APIs into application features, with attention to reliability, latency, and cost.
• Implement embedding pipelines and manage the tradeoffs around model choice, dimensionality, and cost.
• Design agentic and tool-using workflows where they genuinely improve outcomes, and keep them constrained and predictable.
• Evaluate AI features rigorously: build the eval sets and measurements that tell us whether retrieval and generation are actually good, not just plausible.
• Own cost and performance: track token spend, design pricing-aware architectures (including usage and credit models), and keep features economical at scale.
• Integrate AI features into Laravel and Vue applications and the surrounding infrastructure.
• Stay close to the fast-moving ecosystem and bring sound, practical judgment about what is worth adopting.

What We Are Looking For

• 3+ years building production software, with at least 1+ year shipping AI or LLM-powered features that real users touched.
• Hands-on experience with RAG: you have built ingestion and retrieval pipelines, not just called a chat API.
• Solid understanding of embeddings and vector search, including pgvector or a comparable vector store.
• Direct experience integrating LLM APIs into applications, and a realistic sense of their failure modes.
• Strong general engineering fundamentals: APIs, databases (especially PostgreSQL), and writing tested, maintainable code.
• The instinct to measure AI quality rather than trust vibes, and to design evals that catch regressions.
• Cost and latency awareness: you treat token spend and response time as first-class design constraints.
• Clear communication and the judgment to know when AI is the right tool and when it is not.

Nice to Have

• Experience with the Laravel and Vue stack, or strong PHP and modern JavaScript.
• Experience with relevant AI platforms and APIs.
• Familiarity with hybrid search, reranking, and retrieval evaluation techniques.
• Experience designing usage-based or credit-based pricing models for AI products.
• Experience with document-heavy or knowledge-base products (ingestion of varied formats, large corpora).
• Experience deploying on AWS and/or Azure with containers and CI/CD.
• A point of view on agentic and spec-driven development workflows, ideally from real use.
• Experience selling or explaining AI capabilities to non-technical, often skeptical, B2B and industrial buyers.

Why Join

• AI is core to the products, not a bolt-on, so your work ships and matters.
• Real ownership over the AI architecture across multiple products and clients.
• A senior team that respects rigor: they want AI that is measured and reliable, not demos that fall over.
• Remote-first with flexible hours and a focus on outcomes over hours logged.
• A chance to define how a studio builds AI products well, in a domain (industrial and manufacturing) where the opportunity is real and underserved.

How to Apply

Send a short note about your background, an AI feature you shipped to real users (what it did, how you made it reliable, and how you knew it was good), and your view on where retrieval-based AI products are heading.