Resume

ML/NLP researcher focused on practical AI for low-resource languages.

Summary

Tila Muhammad

ML/NLP researcher. Experienced in text classification (sentiment, hate speech, fake news), summarization, and deploying ML apps (Streamlit). Passionate about Urdu NLP and applied AI.

Education

Master of Science in Artificial Intelligence

01/2026 — Present

FAST-NUCES, Islamabad

  • Coursework: Advanced ML, Deep Learning, Natural Language Processing, Understanding of the Holy Quran I & II.
  • Ongoing Project: Cross-Domain Adaptation of Stacking Ensembles for Behavioral Fatigue Detection System
  • Currently designing a stacked ensemble framework for cross-domain behavioral fatigue detection.
  • Focus: Ensemble Learning Domain Adaptation Fatigue Detection

Bachelor of Science in Software Engineering

08/2017 — 09/2021

Islamia College University, Peshawar

Foundation in algorithms, databases, and software engineering; senior projects in NLP and ML.

Professional Experience

AI/ML Intern

11/2024 — 02/2025

Clouditecture, Peshawar

  • Built sentiment models on Yelp (Logistic Regression); 93.27% accuracy, F1 93.27% (P 93.00%, R 93.54%), evaluated on 560k binary-labeled textsCode
  • Built & deployed an email spam classifier (LSTM + Word2Vec) on Streamlit; 83,448 emails (binary), test accuracy 99.00% (val 98.96%), 5 epochsCode
  • Implemented Quora Duplicate Question Detection (KNN) on Quora Question Pairs; accuracy 82.35%, precision 79.86%, recall 73.98%, F1 76.80%Code
  • Automated JD classification → ESCO codes — Code.

DS & AI Trainee

01/2024 — 09/2024

atomcamp, Islamabad

  • News classification on BBC dataset (5 classes); evaluated classical ML models; best: Naive Bayes — 98% accuracy, macro-F1 0.98Code.
  • Student adaptability prediction using behavioral data — Code.
  • CV for weather condition identification from images
  • Urdu discussion summarization with LSTM + K-Means
  • COVID-19 clinical trial analysis — Code.

Odoo Developer Intern

03/2022 — 09/2022

Strings Technologies, Peshawar

  • Customized Odoo ERP modules per business requirements (Sales, Inventory, HR).
  • Diagnosed/resolved core framework issues; improved stability and performance.
  • Maintained technical documentation for custom modules and fixes.

🎓 Certifications & Training

2023

Scientific Paper Writing

Coursera

View
2023

Machine Learning

Simplilearn

View
2025

Bioinformatics Seminar

International

View
2020

Zero to Mastery ML

Udemy

View
2025

IELTS – Band 6.0

L7 · R5 · W6 · S6

2022

KP Digital Internship

Govt. of KPK

View

📚 Research Papers

2025

IoT-Enabled Food Freshness Detection Using Multi-Sensor Data Fusion and Mobile Sensing Interface

IoT Sensors ML

Real-time freshness estimation using pH, moisture, and ethanol sensors. Reduces waste and improves safety.

2023

Hate Speech Detection Using Machine Learning and N-Gram Techniques

NLP Classification Baselines

Classical ML with n-gram features for toxic text detection; a strong, simple baseline for English experiments.

In progress

Transliteration-Invariant Urdu Toxicity Detection with Fine-Grained Labels

NLP Urdu Cross-domain

Robust toxicity detection across scripts and domains using label granularity and transliteration stability.