Case study

OpsMind

Search operational documents and trace answers back to their sources.

Overview

OpsMind helps operators search operational documents, find relevant information, and see the source behind each result.

Problem

Operational data is difficult to retrieve when records are split across systems.

The core problem is retrieval quality: finding relevant operational information and preserving where it came from.

Operational notes live across runbooks, tickets, docs, and incident records

Keyword search misses related records when terminology differs across sources

Engineers need retrieved information with source references, not detached summaries

Solution

A retrieval pipeline for operational documents.

OpsMind separates ingestion, MySQL metadata storage, Redis-backed lookup, and response assembly.

Retrieval-first query path

Queries retrieve source sections first, then assemble a response from the returned records and citations.

Document ingestion pipeline

Documents are parsed, normalized, split into indexed sections, and linked back to source metadata.

Split storage model

MySQL owns structured records and ingestion state. Redis owns the retrieval index for document sections.

Result assembly

Matching sections are collected with source metadata, ranked, and passed into the response step.

Architecture

Backend design decisions.

The architecture keeps durable metadata, indexed retrieval, and request handling in separate responsibilities.

FastAPI backend

Keep ingestion and query workflows behind explicit HTTP endpoints.

FastAPI provides typed request models, clear route boundaries, and a simple async path for backend orchestration.

MySQL database

Use relational storage for document metadata and ingestion state.

Documents, indexed sections, source references, and ingestion status need constraints and inspectable records.

Redis retrieval index

Use Redis for fast section lookup and keep it separate from the metadata database.

The retrieval index can be rebuilt from MySQL-backed source records without becoming the system of record.

Separate endpoint surfaces

Separate write-heavy ingestion from read-heavy query execution.

Ingestion endpoints manage parsing and section creation. Query endpoints handle retrieval and response assembly.

Features

Document ingestion

Indexed retrieval

Query response system

Source attribution

Results

One retrieval path for operational documents

Fast lookup across indexed source material

Source metadata preserved through ingestion and query responses

Note

This is an open-source project demonstrating engineering capability in backend architecture, ingestion pipelines, and indexed retrieval.

View open-source implementation