Skip to content
MJSCRIPT®

Work

Selected Engineering Work

A selection of production systems I've architected, developed, and delivered—spanning distributed business operations, AI-powered automation, and real-time infrastructure.

01  /  TORA & TORY

Architecting an Event-Driven Business Ecosystem

Distributed Systems / Full-Stack Engineering

Separate applications and databases connected through REST APIs, webhooks, and BullMQ jobs backed by Redis, built and owned end to end.

  • React
  • Node.js
  • REST APIs
  • Webhooks
  • BullMQ
  • Redis

Read case study

Conceptual event-driven architecture (mobile layout)Three independent applications with separate databases connect to an integration layer, which feeds Redis queues and BullMQ workers that call back into application APIs.OWN DBApp AREACT · NODEOWN DBApp BREACT · NODEOWN DBApp CREACT · NODEIntegration layerREST + WEBHOOKSRedis queuesBACKED BY REDISBullMQ workersASYNC JOB PROCESSINGREST CALLS BACKWEBHOOK EVENT / JOBREST CALLSEPARATE APPLICATION DATABASE
Conceptual diagram. Illustrates the integration pattern; it is not a verified production topology.

02  /  Mind Matrics Media

Engineering an AI-Powered Advertising Automation Platform

AI Product Engineering / Automation

A Google Ads API web application with AI-agent workflows for campaign creation, performance analysis, keyword discovery, and budget decisions.

  • Node.js
  • React
  • Google Ads API
  • AI Agents
  • API Integrations

Read case study

Conceptual advertising automation loop (mobile layout): ingest, analyze, decide, execute, then back to ingest01IngestGOOGLE ADS API02AnalyzeAI AGENT03DecideBUDGET · KEYWORDS04ExecuteCAMPAIGN CHANGES
Illustrative workflow. No performance data is shown. The loop returns from Execute to Ingest.

03  /  Copartner

Building Real-Time Trading & Streaming Systems

Real-Time Systems / Backend Engineering

React and D3.js trading analytics, Node.js backend services, a reported 30% reduction in backend data latency, and streaming infrastructure for 10,000+ concurrent users.

  • React
  • Node.js
  • D3.js
  • Real-time data processing

Read case study

Stylized real-time data fan-out (mobile layout)Live dataTRADING · STREAMReal-time processingNODE.JS10,000+ concurrent usersCLIENT FIELD IS SCHEMATIC
Stylized interface. Not production telemetry.