Eric Joseph Lee
Back to AI enablement and automation

Omnia Consulting (Freelance)

Production AI Agent Systems

Designed and built three production AI agents for Omnia Consulting clients: a 24/7 FAQ bot and two RAG-powered coaching assistants grounded in 75 real session transcripts. Claude generates the answers, while the transcripts and search index stay on self-hosted hardware with a Cloudflare tunnel, a watchdog process, and documented recovery procedures. These are running systems, not demos.

Role
AI Product Developer & Systems Administrator
Team
Solo build
Timeline
Ongoing — 2026–Present
Year
2026
Production AI agents deployed
3
Production AI agents deployed
Session transcripts indexed (RAG)
75
Session transcripts indexed (RAG)
Availability for clients
24/7
Availability for clients
Diagram: coaching transcripts are chunked, embedded with FastEmbed, and stored in ChromaDB on the client's hardware; Claude answers clients through Telegram, Circle, and a website FAQ agent
1 / 1

Process

  1. 1. Architecture

    RAG pipeline grounded in real organizational content

    • Structured 75 session transcripts (45 minutes each) into a ChromaDB vector store using FastEmbed for embedding.
    • Built multi-step prompt chains that draw from verified organizational content — not generic model knowledge.
    • Designed intake patterns, response formatting, confidence framing, and edge-case handling for each agent.

  2. 2. Infrastructure

    Knowledge base self-hosted on a NAS, answers from Claude

    • Deployed all three systems on a self-hosted NAS with a Cloudflare tunnel for remote access. The transcripts and vector store stay on the client's hardware; Claude processes each request to write the answer.
    • Implemented a watchdog process for automatic crash recovery — maintaining consistent availability.
    • Investigated and documented the full NAS boot chain to resolve persistent auto-start failures — identified volatile RAM rebuild as root cause and established a reliable restart procedure.

  3. 3. Production Bug Fix

    Diagnosed and resolved a message fragmentation issue in production

    • Identified an asyncio message fragmentation bug causing disconnected multi-part replies in the live agent.
    • Implemented a debounce buffer at the code level to restore coherent response behavior.
    • Documented the fix and updated the recovery procedures for future incidents.

Outcomes

  • Three production AI systems running continuously — FAQ agent handling common inquiries automatically, two RAG coaching bots grounded in real client content.
  • The knowledge base stays on the client's own hardware, so the source transcripts are never handed to a third-party platform.
  • Production bug resolved without service interruption — system has maintained reliable operation since.

Skills and tools

  • AI Agents
  • RAG
  • Prompt Engineering
  • Workflow Automation
  • Freelance