Senior Software Architect · Toronto

Backend systems, built to be trusted.

4Mlistings/day
<20mssearch
5production systems
See selected work
01My Profile

Hi, I'm Govia

I've been building software for more than 20 years. The five systems below cover data platforms, applied AI, fintech and media processing. I'm open to full-time roles and contract work.

02Selected Work

Five production systems

01

01North American MLS Data Platform

High-Throughput & Data SystemsPrincipal Architect & Tech Lead
4Mlistings/day8years on the platform300+agents served

Delta ingestion, queue-buffered writes and read-replica clustering that keep search instant while millions of listings change underneath it.

  • MySQL HA Cluster
  • Redis Queue & Cache
  • Microservices
  • Event Pipelines
02

02Licensing Knowledge & Exam Engine

Applied AI & Mission-Critical SystemsSole Architect & Engineer
100+courses1,000+questions10,000+user vocabulary entries

AI-synthesised exam content that stays quarantined until a human domain expert signs it off — then reaches candidates through sub-20ms typo-tolerant search.

  • Offline LLM Pipelines
  • Human-in-the-Loop QA
  • CQRS Dual-Write
  • MeiliSearch
03

03Market Scanner & Dynamic Risk Engine

FinTech & Algorithmic EnginesSystem Architect & Developer
Zeroinbound portsSession-awaredaemonVolatility-adjustedsizing

A resident daemon that replaces brittle cron jobs, then sizes every position against the asset's own volatility before anything is dispatched.

  • Daemon Scheduling
  • Native Math Modeling
  • Risk Arbitrator
  • Webhook Security
04

04Wise Platform API Client SDK

Open Source & API Contract DesignAuthor & Maintainer
Publicon PackagistPest + PHPStanin CIBilingualspec docs

Commands and queries kept apart all the way down to the payment API — the same CQRS split as Case 02, in a domain where a mistake moves real money.

  • CQRS Command/Query Split
  • DTO Contracts
  • RSA/SHA256 Webhook Verification
  • Multi-Driver Auth
05

05Automated Media Processing Pipeline

Media Pipelines & System ToolingLead Systems Engineer
ConcurrentchunkingResumabletransfersIn-memoryremux

A Go daemon that pulls in parallel, remuxes streams in memory without a lossy re-encode, and picks up exactly where a broken transfer stopped.

  • Go Concurrency
  • Stream Remuxing
  • Batch Sync Worker
  • Cloud Storage & CDN
・more projects on GitHub
03Capabilities

Tech stack

01

Backend & High-Throughput

  • Go
  • PHP / Laravel
  • Python
  • MySQL (HA / Read-Replicas)
  • Redis Queue & Cache
  • ElasticSearch / MeiliSearch
  • Microservices
02

Cloud & Infrastructure

  • AWS (ECS / S3 / CloudFront)
  • GCP
  • Docker
  • Nginx
  • Linux System Engineering
  • Webhook Security
03

Systems & Architecture

  • Distributed Data Pipelines
  • Event-Driven CQRS
  • Applied AI / Prompt Chaining
  • Human-in-the-Loop Verification
  • Algorithmic Risk Modeling

Let's connect

Prefer LinkedIn? Message me there, or see what I'm building on GitHub.