Senior AI Engineer for AI Systems

About

We are building AI‑powered features across our product, including intelligent assistants, document understanding, and automation workflows. We are looking for an experienced AI Engineer who has real hands‑on experience building production systems with LLMs and RAG (Retrieval‑Augmented Generation). This is not a prompt engineering role. We need someone who can design and build end‑to‑end AI systems.

Responsibilities

• What you will work on:
  • Build RAG‑based systems using company data (documents, cost catalogs, project data, etc.)
  • Integrate LLMs such as ChatGPT, Claude, and other models into production workflows
  • Design pipelines for:
    • Data ingestion (PDFs, documents, structured data)
    • Embedding and retrieval
    • Context management and ranking
  • Improve response quality, latency, and cost efficiency
  • Implement evaluation and feedback loops for AI outputs
  • Work on real use cases such as:
    • AI estimating assistant
    • Document parsing and insights
    • Smart search across project data
    • AI‑powered recommendations

Requirements (must‑have)

• Strong experience with RAG architectures (vector DBs, embeddings, retrieval strategies)
• Hands‑on experience with LLM APIs:
  • OpenAI (ChatGPT)
  • Anthropic (Claude)
• Experience building production‑grade AI systems (not prototypes only)
• Solid backend skills (Node.js or Python)
• Experience working with structured and unstructured data (PDFs, documents, APIs)
• Understanding of token usage, latency, and cost optimization

Preferred Experience

• Experience with tools like:
  • Pinecone, Weaviate, or similar vector databases
  • LangChain, LlamaIndex (or custom pipelines)
• Experience with document processing (OCR, parsing, chunking strategies)
• Experience with multi‑agent or workflow‑based systems
• Familiarity with AWS or similar cloud environments

What We Care About

• You have built real AI systems used by users
• You think about quality, reliability, and cost
• You can explain trade‑offs (accuracy vs latency vs cost)
• You don’t rely only on frameworks—you understand what’s happening under the hood

Project Scope

• Start with a defined project (2–4 weeks)
• Opportunity for long‑term collaboration if successful

How to Apply

Please include:

1. Example of a RAG system you built (architecture + what problem it solved)
2. How you approach:
   • Chunking
   • Retrieval
   • Context building
3. What you would improve in a typical “basic RAG setup”
4. Links to any relevant projects, repos, or demos

We are looking for someone experienced who can move fast and build reliable systems. If you’ve only experimented with LLMs or followed tutorials, this role is not a fit.