Overview

At Decisiv, I led the transformation of a struggling legacy system into a high-performance fleet management platform. This role showcased my ability to optimize systems under pressure and deliver measurable results.

Major Achievement: 99% Performance Improvement

The Challenge

The existing system was struggling with database query performance, causing slow response times and poor user experience for vehicle asset owners.

The Solution

Built a real-time notifications architecture and completely restructured the database layer:

Metric Before After Improvement
Query Time 8,000ms 70ms -99.1%
Throughput 800 RPM 12,000 RPM 15x
Infrastructure Cost Baseline - 40% reduction

Technical Implementation

  • Query Optimization: Redesigned database queries from the ground up
  • Caching Strategy: Implemented multi-layer caching with Redis
  • Real-time Architecture: Built notifications system using PubNub
  • Frontend: Backbone.js for responsive UI
  • Database: MySQL optimization and indexing strategy

Project: Decisiv Vision

Developed a visualization tool for vehicle asset owners to:

  • Query repair cases with custom search
  • Build personalized view configurations
  • Receive real-time notifications on case updates
  • Access historical data analytics

Technologies Stack

Backend:    Ruby on Rails
Frontend:   Backbone.js, JavaScript
Database:   MySQL
Real-time:  PubNub
Caching:    Redis
Testing:    RSpec, Jasmine

Key Learnings

  1. Performance First: Early optimization decisions pay dividends
  2. Measurement: You can’t improve what you don’t measure
  3. User Experience: Speed directly correlates with user satisfaction
  4. Cost Efficiency: Better performance often means lower infrastructure costs

Impact

This optimization project became a case study within the company for how to approach legacy system modernization. The techniques developed here informed future architectural decisions across the organization.