Back to projects

LLM · Agents · RAG Apr to Jul 2026

Autonomous AI Agent for Network Design Review

Abstract.AI review of network design documents against live device configurations, owned from requirements to operation.

Role
AI Engineer, end-to-end owner
Period
Apr to Jul 2026
Context
ICT systems integrator (via Human Resocia)
Team
About 10
Table 1Results.
18/19 Agreement with reviewer-judged findings; recurring false-positive classes eliminated
25 → 10 LLM calls per review job
min → sec Re-processing time with the content-hash cache
Offline Runs on air-gapped, CPU-only servers

01From problem to approach

The problem

Design reviews meant checking long, multi-format documents against device configs by hand. LLM findings kept repeating the same classes of false positives, and the servers were air-gapped with no GPU.

The approach

Rebuild the fixed pipeline as an agent that plans one step at a time, verify every finding against the source text in code, and run the whole stack offline.

02How it works

  1. Input Design docs + configs Design documents (PDF, PPTX, DOCX, XLSX) and live device configurations.
  2. Step 02 Extract & selective OCR Text, tables, VLAN legends and topology; OCR only on low-quality pages.
  3. Step 03 Offline hybrid RAG BM25 + dense retrieval over ChromaDB, scoped to each device's hostname.
  4. Step 04 Agent plans & acts An LLM planner picks the next tool from a typed registry: decide, act, observe.
  5. Step 05 Guardrails & grounding Code-enforced checks, then verbatim verification of every finding against the source.
  6. Output Review report Grounded discrepancies with evidence, with live progress in Japanese.
Fig. 1Autonomous AI Agent for Network Design Review, step by step.

03What I built

  1. Re-architected a fixed 7-stage LLM pipeline into an autonomous agent: an LLM planner selects tools step by step (decide, act, observe) from a typed tool registry.
  2. Enforced guardrails in code (preconditions, step budgets, a mandatory re-check and a deterministic fallback plan), with graceful degradation when a backend fails.
  3. Built a deterministic grounding layer over LLM findings: verbatim source verification with NFKC / full-width normalization, notation-system rules and section-label device attribution.
  4. Built a multi-format extraction pipeline for PDF, PPTX, DOCX and XLSX, with selective OCR on low-quality pages (Tesseract + OpenCV preprocessing) and python-pptx extraction of tables, VLAN legends and network topology.
  5. Engineered a fully offline RAG stack for air-gapped, CPU-only servers: multilingual sentence-transformers, ChromaDB, hybrid BM25 + dense retrieval, and hostname-scoped filtering to stop cross-device context leakage.
  6. Cut LLM calls per job from about 25 to 10 and added 3-level JSON self-healing; a content-hash cache cut re-processing from minutes to seconds.
  7. Root-caused production asyncio failures (semaphores bound to a stale event loop, dead executors inherited through contextvars) and fixed them at 4 layers.
  8. Served two backends, Gemini / Claude in the cloud and Ollama / Qwen2.5-7B locally, with live Japanese progress updates over WebSockets.

04Technologies

  • Python
  • Django · ASGI
  • LangChain
  • ChromaDB
  • sentence-transformers
  • BM25
  • PyMuPDF
  • python-pptx
  • Tesseract OCR
  • OpenCV
  • PostgreSQL
  • Gemini
  • Claude
  • Ollama · Qwen2.5
  • WebSockets
  • Docker

On request

Request my resume

I share my resume on request. Tell me a little about yourself, and I'll email it to you once I've reviewed your request.

I am a…

Which document?

0 / 100 words

Your details are used only to handle this request and are never shared.