Khawaja
Fashi

Backend Architect/Engineer

I build the part of the app nobody sees — until it breaks. Node.js, Python, and PostgreSQL, shipped async, secured properly, and never requiring a follow-up Slack message.

View My CV
Node.jsNode.js
PythonPython
PostgreSQLPostgreSQL
RedisRedis
DockerDocker
MySQLMySQL
PrismaPrisma
GitGit
LinuxLinux
PostmanPostman
SlackSlack
JiraJira
NotionNotion
Node.jsNode.js
PythonPython
PostgreSQLPostgreSQL
RedisRedis
DockerDocker
MySQLMySQL
PrismaPrisma
GitGit
LinuxLinux
PostmanPostman
SlackSlack
JiraJira
NotionNotion
Node.jsNode.js
PythonPython
PostgreSQLPostgreSQL
RedisRedis
DockerDocker
MySQLMySQL
PrismaPrisma
GitGit
LinuxLinux
PostmanPostman
SlackSlack
JiraJira
NotionNotion
Node.jsNode.js
PythonPython
PostgreSQLPostgreSQL
RedisRedis
DockerDocker
MySQLMySQL
PrismaPrisma
GitGit
LinuxLinux
PostmanPostman
SlackSlack
JiraJira
NotionNotion
// 01About
Khawaja Fashi
BACKEND ENGINEER — 2026

2+ years building robust backend architectures.

Operating with full autonomy across remote and asynchronous environments, I specialize in building high-performance systems in Node.js, Python, and PostgreSQL.

"Slashed backend API response time from ~450ms to under 120ms by rewriting legacy N+1 queries."

// Core Philosophy

Expert in structured written documentation and cross-timezone collaboration. CLEAN CODE + EXPLICIT COMMUNICATION.

CURRENTLY BASED IN ISLAMABAD, PAKISTAN
// 02Professional Experience
May 2026 — Present

Backend Engineer

Quem Systems International

Operating with full autonomy across PKT and EST time zones, architecting scalable infrastructure.

Architected a Redis-based caching layer independently, reducing database read load by 45% and cutting p95 response times by 160ms — designed and shipped asynchronously with zero real-time supervision.

Containerized all backend microservices using Docker, standardizing environment parity across a distributed team and reducing onboarding setup times from 4 hours to 10 minutes.

Extended RBAC middleware to enforce subscription guards on 42 endpoints, independently fixing a critical privilege gap and detailing the resolution via a structured documentation write-up on Notion/GitHub.

Delivered 14+ high-priority features across 4 development sprints entirely async, driving progress across PKT and EST time zones through clean documentation, explicit Slack updates, and thorough PR reviews.

// 02Professional Experience
March 2026 — May 2026

Backend Intern

Quem Systems International

Optimized legacy systems and implemented security protocols for mission-critical infrastructure.

Slashed backend API response time from ~450ms to under 120ms (73% reduction) by rewriting legacy N+1 MySQL queries and generating optimal composite indexes via Prisma ORM independently.

Engineered a 4-tier security framework (HS256 JWT) securing 113 routes across user, admin, and subscription roles with bulletproof cookie isolation and account-state verification.

Accelerated admin dashboard workflows by batching 17 queries via Promise.all, mitigating UI blocking to render user metrics and multi-platform social data in one unified network round-trip.

Implemented an Axios interceptor managing silent JWT rotation; self-documented the architecture to enable seamless async integration for frontend developers without needing synchronous meetings.

// 03Active Projects
Cloud-Native Architectures

IoT Rogue Traffic Simulator

Microservices Migration

HypeWait Queue Drops

Scalable Systems Design

LiveTeamGames Grid

#Python#MQTT#Scikit-learn#Streamlit

A real-time anomaly detection pipeline that flags rogue traffic patterns using a trained Random Forest classifier.

95%+Anomaly Accuracy
<300msModel Latency
350K+Trained Entries

Built an end-to-end MQTT telemetry ingestion pipeline using Python to parse data across 20+ simulated IoT points.

Trained a Random Forest model on 350K+ entries, optimizing classification overhead to keep processing speed under 300ms.

Reduced mean detection-to-alert latency from 5 minutes to 15 seconds using live database-driven dashboards.