Work Education Experience Skills Contact
Available for remote work

Abdullah
Khalid

AI Engineer

I build production AI systems — LLM pipelines, distributed backends, real deployments on GCP. Not notebooks. Not demos. Things that run under load and stay up.

View my work GitHub  ↗
Abdullah Khalid — AI Engineer

2 years shipping production AI  ·  FastAPI  ·  LLMs  ·  GCP  ·  Redis  ·  Currently building Selfimoji @ CoreX Tech

Selected Work

Things I've shipped

Production systems with real users and real load — not portfolio demos.

Production

Selfimoji

AI image & video generation platform for iOS. Users generate personalised photos and videos across different expressions, styles, and scenarios, create video ads, try on clothes virtually, and share via a native iOS keyboard extension — usable in iMessage, WhatsApp, or any messaging app — all powered by per-feature AI pipelines I built from scratch.

FastAPIRedisGCP Cloud Run Infrastructure as CodeSwift / iOS CI/CDLLM PipelinesJob Queue System
Per-feature AI pipelines Distributed job queue Apple IAP backend iOS keyboard extension UGC — image & video Virtual clothes try-on

Integrated Kling AI, Google Veo, and ElevenLabs into chained generation pipelines — with a Redis job queue handling retry, deduplication, and locking so no generation task loses state. Full GCP infrastructure managed with Terraform and deployed via Cloud Build on every commit.

Production

Email Sequencing

Multi-tenant B2B outreach platform with a 3-process distributed architecture — API server, background worker, and React Native mobile app. Handles the full campaign lifecycle: AI-generated per-contact email sequences, scheduled delivery, open tracking, and dual-mode reply detection via IMAP and Gmail Pub/Sub.

LLM APIsNode.jsExpress.jsTypeScriptReact Native
Multi-tenant architecture 7 LLM providers supported PostgreSQL job queue AI persona + enrichment engine

Worked on the PostgreSQL-backed job queue with idempotency keys, retry logic, and per-step scheduling — no external queue dependency, no duplicate sends. Also owned the Gmail Pub/Sub reply detection layer, deployed alongside a React Native mobile client via Expo EAS.

Shipped

PhysioGo

Real-time physiotherapy app using MediaPipe pose estimation. Tracks reps, detects form errors in live video, syncs session data between physiotherapists and patients.

MediaPipePythonFlutterFirebaseDocker
View on GitHub  ↗
Shipped

Grounded

Full RAG pipeline querying 1,000+ pages with accurate answers in under 2 seconds. FAISS vector retrieval with a React frontend — reduced manual search time by 70%.

<2s response time 70% less manual search
FastAPIFAISSLangChainReactEmbeddings
View on GitHub  ↗
Shipped

Copenhagen Bicycle Market

Collected and processed 1,000+ records via scraping and preprocessing. Built Regression and SVM models to forecast seasonal demand and analyse pricing trends.

scikit-learnSVMRegressionPythonWeb Scraping
View on GitHub  ↗
Shipped

NLP Pipeline

Urdu sarcasm detection, English-Urdu machine translation using RNN/LSTM, and Phi-3 fine-tuning. End-to-end pipeline from preprocessing through classification and evaluation.

Phi-3 Fine-tuningRNN / LSTMNLPPythonscikit-learn
View on GitHub  ↗
Shipped

AI Exam Seating System

K-Means clustering to group 2,500+ students by domain and batch. Generates automated PDF seating plans and faculty assignments for university exams — zero manual allocation.

K-MeansClusteringPythonPDF Generation
View on GitHub  ↗
Shipped

Pedestrian Gender Detection

Binary image classification using HOG, LBP, GLCM and VGG19 deep features. PCA dimensionality reduction with Linear SVM, evaluated via 10-fold cross-validation.

VGG19HOG / LBPSVMPCAComputer Vision
View on GitHub  ↗

Education

Where I studied

Aug 2021 — June 2025
FAST National University (NUCES)
Bachelor of Science in Computer Science
  • Graduated from one of Pakistan's top CS programs
  • Final year project: PhysioGo — AI-powered real-time physiotherapy tracking system
  • Coursework included AI, Machine Learning, NLP, Computer Vision, and Systems Design

Experience

Where I've worked

Dec 2025 — Present
CoreX Tech
AI Engineer
  • Designed and deployed a scalable FastAPI backend with auth, webhooks, and async workflows via Redis
  • Built a distributed job system with retry, locking, and idempotency — AI tasks never lose state under concurrent load
  • Implemented full CI/CD on GCP (Cloud Build + Cloud Run) with Infrastructure as Code — deploy on every commit
  • Built Apple IAP backend managing subscriptions, free trials, and consumables via JWT-secured webhooks
  • Developed Email Sequencing system using LLM APIs, cutting manual outreach effort by 90%
Aug 2025 — Nov 2025
Technupur (Pvt) Ltd
Software Engineer Intern
  • Built features for a production SaaS application, improving reliability and user experience
  • Integrated REST APIs with structured error handling, reducing failure rates and improving data consistency
  • Optimised UI performance by eliminating unnecessary re-renders and redundant API calls

Capabilities

What I work with

AI & Machine Learning
LLMsRAG LangChainFAISS Fine-TuningEmbeddings Vector DatabasesNLP Pose EstimationPinecone
Backend & Infrastructure
FastAPINode.js FlaskRedis PostgreSQLMongoDB DockerGCP Cloud Run CI/CDIaC
Languages & Frontend
PythonTypeScript JavaScriptSwift C++Java ReactNext.js FlutterReact Native

Contact

Let's build
something real.

Open to remote freelance projects and full-time roles at ambitious companies. If you need AI systems that work in production — not just in a notebook — let's talk.

✉  Send an email LinkedIn  ↗ GitHub  ↗ +92 311-7223011  ↗