M

Muhammad Mueez Rizwan

About

Detail

Islamabad, Pakistan

Timeline

work
Job
school
Education
folder
Project

Résumé

Jobs verified_user 0% verified
  • Ptcl
    Data Analyst Intern
    Ptcl
    Jul 2025 - Sep 2025 (3 months)
    • Partnered with cross-functional teams to maintain 100% data integrity across operational reporting pipelines • Designed and executed complex SQL queries on large datasets, reducing data retrieval time by 40% and enabling faster decision-making • Proposed an ML-based network anomaly detection framework adopted in a post-internship pilot, demonstrating applied AI problem-solving on production data
Education verified_user 0% verified
  • IBM
    Python for Data Science, AI & Development
    IBM
    Aug 2024 - Sep 2024 (2 months)
  • S
    Introduction to SQL
    Simply Learn
    Jun 2024 - Jul 2024 (2 months)
  • N
    Bachelor of Science in Data Science
    National University of Computer & Emerging Sciences
    Jan 2022 - Jan 2026 (4 years 1 month)
Projects (professional or personal) verified_user 0% verified
  • Personal Project
    Cosmic TaskPro – Full-Stack Task Management System
    Personal Project
    Jan 2026 - Current (8 months)
    • Built a high-performance RESTful API using FastAPI and Pydantic V2, implementing JWT-based authentication and Google OAuth for secure user session management across 2 auth flows • Engineered an automated background notification engine using Python BackgroundTasks and SMTP, dispatching real-time email alerts to users 1 hour prior to task deadlines with 99% delivery reliability • Architected a timezone-agnostic data pipeline ensuring 100% accurate synchronization between browser-side inputs and SQLite storage, eliminating global time-shift discrepancies across all user sessions • Designed a responsive Glassmorphism UI using Vanilla JavaScript and CSS3, achieving sub-200ms page load times with a focus on modern UX principles
  • f
    LullabAI – AI-Powered Animated Storytelling Platform
    final year project
    Jan 2026 - Current (8 months)
    • Architected full-stack AI storytelling platform serving 100+ daily users, integrating a multi-modal pipeline across LLM, image synthesis, and TTS modules • Engineered Retrieval-Augmented Generation (RAG) system using Claude-3 with vector-indexed story corpus; achieved 93% prompt-to-output relevance and enforced age-appropriate vocabulary via custom system prompt constraints • Optimised Stability AI video synthesis pipeline through async batching and prompt compression, cutting generation latency by 60% while maintaining 1080p quality • Integrated Google Cloud TTS with dynamic provider fallback, achieving 99% narration success rate across 4 regional accent configurations
  • Personal Project
    AI Chatbot for Retail Shop
    Personal Project
    Jan 2026 - Current (8 months)
    • Designed and deployed a custom conversational AI assistant for a local business, handling product queries, order status, and FAQs from a domain-specific knowledge base of 800+ entries • Engineered a structured prompt system with role-definition, few-shot examples, and output guardrails; reduced hallucination rate from 22% (baseline) to under 4% on domain-specific queries • Integrated inventory lookup via REST API, enabling real-time stock availability responses with average end-to-end latency under 1.2 seconds • Achieved 87% user satisfaction rate in 2-week post-deployment evaluation based on 150+ logged and reviewed conversations
  • A
    Real-Time Fake News Detection Pipeline
    Academic Project
    Jan 2025 - Dec 2025 (1 year)
    • Fine-tuned a BERT-based transformer classifier on 120K+ news articles, achieving F1-score of 0.91 on held-out test set across 4 misinformation categories • Built drift-aware MLOps retraining pipeline; drift-triggered strategy reduced model performance degradation by 34% versus periodic retraining baseline • Containerised full pipeline with Docker Compose (6 services); configured Prometheus scraping and Grafana dashboards for real-time model health monitoring • Implemented CI/CD via GitHub Actions with automated unit tests and model validation gates, reducing deployment failures from 3 incidents per week to zero over a 30-day window