Muhammad Bilal Akram

Highly motivated Software Engineering graduate (COMSATS University, 2026 | CGPA: 3.48) with 2 industry internships and 10+ shipped projects spanning full-stack web, machine learning, and cross-platform mobile development.

At Developerhub

Corporation, single-handedly delivered 2 production web applications, a multi-vendor e-commerce platform and a real-time entrepreneur-investor network from zero to live deployment.

At Arch

Technologies, trained and deployed 4 ML models (spam classifier, digit recognizer, flower classifier, House price prediction model) with model explainability using LIME and SHAP.

Final Year

Project (VendoMart) integrates Gemini AI, real-time WebSockets, Stripe payments, and a full React Native web and mobile app across a 3-sided marketplace. Fluent across Python/Django, Node.js/Express, React Native, PostgreSQL, Scikit-learn, and TensorFlow and consistently productive at the intersection of all three.

Targeting Junior Software

Engineer, Backend Developer, ML Engineer, or React Native Developer roles.

Full-Stack Software Development Intern - Developerhub Corporation

(2025-07 - 2025-08)

• Engineered MarketVibe, a production-grade multi-vendor e-commerce platform serving 3 distinct user roles (admin, seller, buyer) with 8+ core modules, Stripe Connect payments, dynamic product catalog, smart search, gift campaigns, quote system, and order management deployed live on PythonAnywhere.
• Architected CoFound, a real-time professional networking platform connecting entrepreneurs and investors, featuring bidirectional WebSocket messaging (Django Channels + Daphne), meeting scheduling, fund-raising feeds, and JWT-secured RBAC reducing manual networking friction to a single platform.

Machine Learning Intern - Arch Technologies

(2025-08 - 2025-09)

• Trained an Email Spam Classifier using Python and Scikit-learn with TF-IDF vectorization, achieving multi-model benchmarking across 3+ classifiers and integrating LIME explainability to surface per-prediction feature importance; packaged as an interactive Streamlit application.
• Engineered an MNIST Digit Recognition pipeline comparing Logistic Regression against a Convolutional Neural Network (CNN) in TensorFlow/Keras on 70,000 training samples, deploying a real-time Streamlit app that accepts live drawn-digit input for instant inference.
• Constructed an Iris Flower Classification system covering the full ML lifecycle across 3 species and 150+ samples — data preprocessing, feature engineering, cross-validated model selection, and rich SHAP + Plotly + Seaborn explainability dashboards.

Bachelor of Software Engineering - Software Engineering - COMSATS University Islamabad — Sahiwal Campus (2022-09 - 2026-06)