Babur Shahbaz
2+
Years Exp
10+
Projects
5
Industries
5+
AI Models
Available for opportunities

ML ENGINEER! I'M BABUR

I'm a Machine Learning Engineer who designs, trains, and deploys AI systems that power smarter products and faster decisions.

2+ years building end-to-end ML pipelines and shipping AI-powered products — across telecom, finance, operations, sales, education, and production AI systems.

Currently:  |

View My Work & Get In Touch

View Projects  → Contact Me

How Can I Help You?

🧠

ML Model Development & Deployment

I build, train, and deploy high-accuracy machine learning models that go far beyond analysis — they power real products. Whether it's churn prediction in telecom or multimodal emotion AI, I engineer production-grade pipelines that keep you ahead.

⚡

AI Systems & API Engineering

A brilliant model locked in a notebook helps nobody. I build FastAPI backends, connect trained ML models to live applications, and ship them to the cloud. I turn your AI idea into a running product that delivers results in real time.

🔬

Applied AI Research & Optimization

With hands-on experience across multiple domains and industries, I help teams pick the right architecture and apply the right techniques. I research, prototype, and fine-tune AI systems that improve measurable outcomes at scale.

About Me
Babur Shahbaz Khokhar
Top 1% MS Batch

Research Meets Real World

I turn cutting-edge ML research into production-ready systems that actually ship.

Specializing in deep learning, NLP, and generative AI — my MS thesis built a live multimodal emotion detection system, now deployed on HuggingFace. Ranked top 1% of my MS batch at Bahria University.

Profile Machine Learning & AI Engineering
Education MS Data Science — Bahria University  |  3.75 CGPA
ML Stack PyTorch  ·  HuggingFace  ·  FastAPI
Email baburshahbaz8@gmail.com
10+ Projects Completed
LinkedIn  →
🤗

HuggingFace Deployed

Live Sentira AI Model

→
🔥

PyTorch & HuggingFace

ML Frameworks · Model Fine-tuning

→
🎓

Top 1% MS Batch

CGPA 3.75 · Bahria University

→

Work Experience

June 2024 — June 2025

ML & Data Analyst

Safeena Telecom — Oman
  • Built Power BI pipelines to track and optimize regional marketing campaign performance.
  • Engineered SQL churn prediction models to detect and flag high-risk customer segments early.
  • Automated end-to-end data pipelines for fast, daily ML-ready reports to senior leadership.
  • Turned raw datasets into clear, ML-driven strategic insights to support CMO decisions.
  • Maintained data pipeline integrity in secure, virtualized cloud environments.
March 2021 — March 2024

Data Engineer & ML Developer

Punjab Education Foundation — Pakistan
  • Built and optimized ETL pipelines across 36 districts, cutting reporting time by 40%.
  • Trained ML models to detect at-risk students and trigger early academic interventions.
  • Designed live SQL-connected dashboards to monitor key performance indicators in real time.
  • Built data cleaning pipelines to ensure 100% accuracy across large education datasets.
  • Turned ML model outputs into clear, actionable field-level policy recommendations.

AI & Technical Spectrum

Python (PyTorch · NumPy · Scikit-learn)95%
Deep Learning & Neural Networks95%
Machine Learning & Model Training92%
NLP & Transformer Models (HuggingFace)90%
MLOps & API Deployment (FastAPI)88%
SQL & Data Engineering85%
Generative AI & RAG Systems83%
Agentic AI & LLM Pipelines80%

Education

2024 — 2026

MS Data Science

Bahria University, Islamabad
  • Specializing in AI, ML, DL, ANLP, Big Data and Cloud Computing.
  • Thesis: Multimodal Emotion Detection using EEG, Audio and Video through Adaptive level Fusion
  • Consistently ranked in the top 1% of the batch for academic excellence.
  GPA: 3.75 / 4.00
2014 — 2018

BS Computer Science

COMSATS University, Islamabad
  • Strong foundation in Algorithm Design and Software Engineering principles.
  • Focused on Database Management and Complex Computational Mathematics.
  • Awarded First Class Distinction for high performance in capstone projects.
  First Class Distinction
Portfolio

My Projects

Real-world Machine Learning & AI Engineering projects spanning model deployment, NLP systems, deep learning, and production AI pipelines.

SENTIRA — Multimodal Emotion Recognition
01
⭐ Featured — MS Thesis
Try Live Demo
Multimodal AI Audio · Video · EEG Confidence-Weighted Fusion

SENTIRA Multimodal Emotion Recognition — Audio + Video + EEG Fusion

A tri-modal emotion recognition system fusing speech, facial video, and EEG signals into one confidence-weighted prediction. Combines a Wav2Vec2 + Whisper + RoBERTa audio model, an MS-TAP video ensemble (ViT + landmarks + optical flow), and a multi-stream EEG model — deployed end-to-end as a production API with a live web frontend.

99.50% Fusion Accuracy
3 Modalities
5 Emotions
PyTorch Wav2Vec2 FER-ViT EEGNet FastAPI
Demo Code
AI ATS Resume Screening
02
🤖 AI NLP
Try Live Demo
NLP Naive Bayes TF-IDF

AI ATS Resume Screening System HR Intelligence Tool

Upload 50–100 PDF resumes and a job description — AI ranks every candidate with ATS scores using TF-IDF cosine similarity, Naive Bayes classification, and multi-factor skill matching. Exports ranked results to Excel.

100+ Resumes
4 Job Classes
Live Deployed
Python FastAPI scikit-learn HuggingFace
Demo Code
AI Voice Emotion Detection
03
Try Live Demo
Deep LearningAudio AIPyTorch

AI Voice Emotion Detection System MS Thesis Project

Built a real-time AI system that detects human emotions (happy, sad, angry, neutral, calm) from audio using deep learning. Supports both live recording and file upload.

99% Accuracy
5 Emotions
Live Deployed
PythonWav2Vec2WhisperVercel
Demo Video Code
ProductPulse AI
04
📊 New
Try Live Demo
RAG NLP Product Analytics

ProductPulse AI RAG-Powered Product Intelligence

Upload App Store reviews, support tickets, or surveys — AI extracts top pain points, feature requests, and sentiment in seconds. Ask questions in English or Roman Urdu. Exports a full Excel intelligence report. 100% free stack.

RAG Pipeline
EN+UR Bilingual
Free Zero Cost
FastAPI Mistral SentenceTransformers Vercel
Demo Code
Video Emotion Detection
05
🎥 New
Try Live Demo
Computer Vision Deep Learning OpenCV

Real-Time Video Emotion Detection Facial Emotion AI System

Built a real-time facial emotion recognition system using deep learning and computer vision. The system detects emotions directly from webcam video streams with live prediction and confidence scoring.

Real-Time Inference
5+ Emotions
Live Web Demo
Python TensorFlow OpenCV Flask
Demo Code
TalentAura AI Interview System
06
🎙️ New
Try Live Demo
Generative AI Speech AI Emotion AI

TalentAura — AI Interview System Adaptive · Multilingual · Emotion-Aware

Avatar-based AI interviewer (ARIA) that parses your CV, verifies identity via face matching, asks adaptive questions in English or Roman Urdu, detects emotions from voice in real-time using Sentira, scores every answer with NLP, and exports a full Excel report.

5 Emotions
Real-Time Adaptive
Live Deployed
Whisper DeepFace FastAPI Sentira
Demo Code
Telecom Churn Dashboard
07
View Project
Power BIDAXArcGISSQL

Telecom Churn & Retention Dashboard

End-to-end Telecom Business Analysis using Power BI with advanced DAX measures and ArcGIS map visuals to identify top 5 highest-risk cities and churn patterns.

Top 5Risk Cities
MultiFilters
LiveDashboard
Power BIDAXArcGIS
View
Credit Card Business Analysis
08
View Project
Power BIDAXPostgreSQLForecasting

Credit Card Business Analysis & Insights

Analyzed customer spending behavior by income, gender, and card type. Revealed high-income men spent 3× more; Blue cards generated 46× more revenue than other types.

$55.3MRevenue
655K+Transactions
46×Revenue Diff
Power BIPostgreSQLDAX
View
Explore All Projects on GitHub  

Get In Touch

Reach out for Machine Learning Engineering, AI development, or Data Science opportunities

Location

Islamabad, Pakistan

Phone

0330-0399995

Email

baburshahbaz8@gmail.com

Resume

Download CV

Have a Project?

Let's connect and explore how we can bring your next ML product or AI system to life.

Send Me an Email