Open to internships & collaborations

AI UNDERGRADUATE · DEVELOPER · DESIGNER

I build practical AI products and thoughtful digital experiences.

I’m Shayan Akbar, a Bachelor of Artificial Intelligence student at the University of Wah. I work across Python, machine learning, intelligent applications and product-focused UI/UX—turning ideas into things people can actually use.

Explore my work ↘ Get in touch

Current focusAI + ML systems

Based inWah Cantt · Pakistan

3.64Current CGPA

3AI / ML internships

8Public GitHub repos

2029Expected graduation

PYTHON✦MACHINE LEARNING✦STREAMLIT✦TF-IDF✦SQL✦JAVA✦FIGMA✦GIT / GITHUB✦ PYTHON✦MACHINE LEARNING✦STREAMLIT✦TF-IDF✦SQL✦JAVA✦FIGMA✦GIT / GITHUB✦

01 · ABOUT

Building foundations that compound.

I started building my technical portfolio early because I’d rather learn by shipping than wait for a degree to make the work feel “official.”

My current path combines machine learning fundamentals with software development, deployment and interface design. The goal is simple: become the kind of AI engineer who understands not only the model, but the product around it.

I’m currently strengthening deep learning, NLP, computer vision, generative AI and production-ready application development.

Read my full CV ↗

Education

BS Artificial Intelligence

University of Wah

2025 — 20293rd semester

Current direction

AI Engineering

Machine learning · Deep learning · NLP · Computer vision

Working principle

Learn → Build → Test → Document → Deploy

02 · SELECTED PROJECTS

Work that shows how I think.

A mix of deployed applications and hands-on AI experiments. I prioritize explainable logic, clean structure and usable interfaces.

01Deployed application

TechMatch AI — Tech Stack Recommender

A transparent content-based recommendation system that maps user skills and career goals into a shared TF-IDF space, scores roles with cosine similarity, and explains the strongest matches and skill gaps.

JavaScriptTF-IDFCosine SimilarityGitHub Pages

Live app ↗ Source ↗

02Python application

Nova — Rule-Based AI Chatbot

A deterministic conversational application built around input normalization, explicit rules, fallback handling and continuous interaction—without pretending to be an LLM.

PythonControl FlowChatbot Logic

View repository ↗

03Deep-learning experiment

Music Generation AI

An LSTM-based experiment that learns patterns from MIDI compositions and generates new playable musical sequences.

PythonLSTMKerasMIDI

View repository ↗

04Machine-learning case study

Credit Card Fraud Detection

Explored supervised and anomaly-detection approaches for highly imbalanced transactions, with emphasis on threshold selection, precision/recall trade-offs and explainability.

Random ForestXGBoostSMOTEEvaluation

Browse all GitHub repositories ↗

03 · EXPERIENCE

Learning through real deliverables.

Jul — Aug 2026

Artificial Intelligence Intern · DecodeLabs

Built a rule-based chatbot and a content-based tech-stack recommender; practiced structured Git/GitHub workflows, testing, documentation and deployment.

May — Jul 2026

Machine Learning Intern · Live Pakistan

Worked on customer churn, fraud detection, data preparation, classification and comparative model evaluation with attention to class imbalance and explainability.

May — Jun 2026

Artificial Intelligence Intern · CodeAlpha

Completed practical AI assignments spanning translation workflows, FAQ chatbot logic and an LSTM-based music-generation experiment.

04 · UI/UX & PRODUCT DESIGN

Engineering mindset, visual discipline.

Self-initiated SaaS dashboard concepts built to practice information architecture, responsive layouts, reusable components and portfolio presentation.