HASNAINMUAVIA
Based in Lahore, Pakistan,
I'm a developer and
AI/ML engineer.
About Me
"You can't use up creativity. The more you use, the more you have."
Passionate AI/ML Engineer & Full-Stack Developer
Proven AI/ML Engineer and Full-Stack Developer with expertise in backend development using Django, Flask, and FastAPI. Published researcher with two peer-reviewed papers in MDPI on AI-generated content detection and visual cyberbullying. Specialized in building RAG applications, generative AI solutions, intelligent automation agents, and production-level systems. Successfully deployed multiple production-ready platforms including AKTI Management Software and POS systems. Experienced AI/ML instructor having conducted 9+ training batches and delivered 17+ professional courses, training industry professionals and students at leading institutions including NUST, FAST, UMT, and PUCIT. I am dedicated to bridging the gap between theoretical AI concepts and practical, scalable software solutions, constantly exploring new technologies to drive innovation and efficiency.
9+
Training Batches
30+
Projects Done
17+
Courses
50+
Interns Trained
My Journey
A timeline of building, leading, teaching, and shipping AI products.
13+
Roles
5+
Companies
350+
Mentees
Development Team Lead
Nexxaura (Part-time)
- Leading a development team to build intelligent automation solutions while managing agile sprint cycles and delivery planning
- Coordinating development tasks, mentoring team members, and ensuring high-quality, scalable software delivery
- Leading training for the development team on an Agentic Medical Billing Automation System and AI-driven healthcare workflow automation
AI & Tech Architect
Wood Tech Mobel (Part-time)
- Designing and developing scalable CRM systems enhanced with AI-driven automation and modern software architecture
- Implementing advanced AI solutions including RAG for contextual data access and ElevenLabs-powered calling agents for real-time interactions
- Building intelligent workflow automation pipelines with n8n to streamline operations and improve decision-making
Generative AI Mentor
The University of Chenab, Gujrat (Contract)
- Delivered practical AI training to a cohort of 350 students, with a strong focus on hands-on learning outcomes
- Taught agentic AI bots, AI development, website development, and AI-powered content generation through project-based sessions
- Mentored students on real-world projects to build industry-ready skills in automation, intelligent systems, and digital product development
Development Team Lead
BuiltPulse
- Leading technical infrastructure and core systems engineering while defining long-term technology strategy
- Managing and mentoring cross-functional distributed engineering teams across Dubai, Karachi, and Lahore
- Overseeing architectural design, technology stack selection, and technical roadmap planning for scalable systems
Software Engineer - Gen AI
BuiltPulse
- Designed and maintained scalable backend systems using Django across 14 group companies
- Built and deployed Agentic AI agents to automate complex business processes and decision-making workflows
- Integrated AI-driven automation with existing systems, APIs, and databases for seamless end-to-end workflow automation
Lead Automation Mentor
Tecrix
- Conduct AI and Automation workshops at universities, training students on n8n, Zapier, and Make.com
- Design project-based learning experiences focused on real-world automation and low-code development
- Helped students build portfolio-ready automation projects and transition to practical implementation
Associate AI/ML Engineer
Big Binary Tech (Remote)
- Developed and deployed AI-powered platforms including POS systems, portals, and invoicing solutions using Django, FastAPI, and Flask
- Designed and implemented email automation agents and RAG systems to enhance system intelligence
- Built advanced AI automations including email agents, workflow systems, and intelligent assistants
Associate Software Engineer
Arfa Karim Technology Incubator, DHA Campus
- Built and deployed POS, portal, email agent, AI tutor, and invoicing platforms with scalable backend architecture
- Delivered end-to-end software features from planning and development to testing and deployment
- Improved system performance and reliability through optimized backend workflows and modular code structure
Gen-AI Engineer (Hybrid)
Autometa Limited
- Developed AI-powered automation solutions including intelligent agents for task automation
- Built RAG system integrating OpenAI and Anthropic models with LangChain and Model Context Protocol (MCP)
- Implemented MCP integrations with JIRA, Slack, and GitHub for seamless workflow automation
AI and Web Instructor
Arfa Karim Tech Incubator (DHA), Lahore, Pakistan
- Conducted 9+ training batches and delivered 17+ professional courses in Generative AI, Agentic AI, Advanced Python, Web Development, and Django
- Delivered on-site training at Allied Schools, The Punjab School, UMT, NUST, FAST, and PUCIT
- Led and hosted multiple online webinars focused on Generative AI and AI-driven automation
Team Lead
Arfa Karim Tech Incubators - DHA Franchise
- Led and delivered Web, Python, and AI internship programs
- Designed international-standard curriculum aligned with industry demands
- Conducted technical interviews and contributed to hiring decisions
Generative AI Course Trainer
Arfa Karim Tech Incubator (ASTP)
- Taught Full-Stack Web Application development using Generative AI, automation, and integration
- Guided students in Deep Learning, Machine Learning, and NLP with hands-on projects
- Provided industry-focused AI training to enhance real-world application skills
NAVTTC Advanced Python Trainer
Arfa Karim Technology Incubator, DHA Campus
- Trained NAVTTC students in Generative AI, Machine Learning, Deep Learning, and Django
- Conducted a 15-day internship program for two NAVTTC batches
Featured Works
AI/ML & Deep Learning
AI-Generated Cryptocurrency Tweet Detection
Context-aware representation learning model to detect AI-generated cryptocurrency tweets across multiple LLMs. Achieved 99% accuracy. Published in MDPI 2025.
Multi-Class Visual Cyberbullying Detection
End-to-end deep learning pipeline for visual cyberbullying detection using TensorFlow and Computer Vision. Achieved 98% accuracy. Published in MDPI 2024.
Gastrointestinal Endoscopy Predictions
Multi-class classification system for 11 gastrointestinal conditions using VGG-16, VGG-19, DenseNet121. Best accuracy: 97% with VGG-16.
Lung and Colon Cancer Prediction
Transfer learning with layer freezing for medical image classification on LC25000 dataset. Achieved 99% accuracy.
Bone Fracture Prediction
Real-time hospital dataset with 8 bone fracture classes. Implemented VGG-16, VGG-19, ResNet50v2, DenseNet models.
MNIST Handwritten Digit Prediction
Neural network model to recognize handwritten digits from MNIST dataset. Achieved 90.2% accuracy.
Election Prediction - 2024
Scraped election data, performed EDA, trained ML model to predict election outcomes.
IMDB Movies Sentiment Analysis
Scraped IMDb reviews using Selenium, preprocessed with NLTK, displayed sentiment analysis on Gradio using VADER.
RAG Systems & Generative AI
AccountingBot RAG System
Intelligent RAG system with web scraping via Selenium, vector embeddings using LangChain, authentication, and session management.
EduAI - Learning System
AI-powered educational platform using Groq, LangChain, Llama Vision, RAG, and OCR. Features Lecture AI, Book Chat, and quiz generation.
Story Crafter - Comic Book Creator
AI agent that generates comic book-style images based on story context with consistent character design and scene progression.
JIRA Issue Creator Tool
AI agent that extracts tasks from project priorities and automatically creates JIRA issues through function calling.
LearnOCR - AI Textual and Vision Tool
Integrated Groq API with multiple models, Tavily Search for retrieval, and Llama-Vision for image insights.
LLAMAENG - Data Preprocessing Tool
AI tool for data preprocessing and labeling using Llama models. Context-aware word correction feature.
VisionAID Hackathon Project
AI-powered medical image analysis tool with IBM Watson, Aria API, TTS/STT for voice-based interaction.
RAG System With Llama-Index
Robust RAG system using Llama3.1:8B local model with runtime document embedding and Streamlit interface.
Full-Stack Django Projects
PON-P3 Pathogenicity Prediction
Bioinformatics web app for Lund University to predict amino acid variation pathogenicity using Django, Celery, RabbitMQ.
AKTI Management System
Comprehensive management system with portal and POS on Vercel with Neon PostgreSQL. Role-based access control, analytics, invoicing.
Transcripto
Transcription app supporting audio, video, and YouTube links with Whisper AI and Groq Llama 3.2 Q&A system.
PON-GOF for University of LUND
Database validation with 200 bootstrapped models for prediction. Generates CSV reports sent via email.
Dynamic Portfolio with Django
Fully responsive portfolio with email configuration, password reset, and inline admin panel with runtime color changing.
Product Price Comparison
Django app to scrape and compare product prices from three online stores using Selenium.
Flask & Other Projects
Accounting RAG Bot (Flask)
RAG bot for accounting queries using Llama3.2 and Groq API with GPT-like interface and admin upload page.
Analytics Dashboard
Dashboard for monitoring user queries and reviewing bot responses with chat history saved to database.
Flask CRUD API
CRUD API using Flask with MVC architecture for clean and organized codebase.
Selenium Picture Downloader
Automation script to download pictures from Unsplash with dynamic content handling.
Education
Bachelor of Software Engineering (BSSE)
University of Management and Technology (UMT), Lahore, Pakistan
F.Sc (Pre-Engineering)
Superior College, Lahore, Pakistan
Skills & Expertise
Production-tested across 6 disciplines — from transformer fine-tuning to scalable Django backends.
AI & Machine Learning
17 Skills
Full-Stack Development
17 Skills
Data Engineering & Analytics
11 Skills
Cloud & Databases
9 Skills
Tools & Platforms
27 Skills
Analytical Skills
4 Skills
Research Publications
Published in MDPI journals — exploring AI-generated content detection and visual cyberbullying with novel deep-learning architectures.
A Context-Aware Representation-Learning-Based Model for Detecting Human-Written and AI-Generated Cryptocurrency Tweets Across Large Language Models
Journal of Risk and Financial Management (JRFM), MDPI
This study addresses AI-generated financial text detection by constructing a dataset of financial tweets regenerated using six LLMs. A context-aware DeBERTa model was fine-tuned and compared against other transformers and traditional ML models. The proposed model achieved 94% accuracy, precision, recall, and F1-score, significantly outperforming competitors and demonstrating the effectiveness of context-aware representation learning for text authentication.
Multi-Class Visual Cyberbullying Detection Using Deep Neural Networks and the CVID Dataset
Information, MDPI
This paper presents a novel dataset (CVID) for visual cyberbullying detection with four classes: abuse, curse, discourage, and threat. Using deep learning models including VGG16, VGG19, and Vision Transformer, the proposed DenseNet201-based model achieved 99% test accuracy. The 5-fold stratified K-fold validation confirmed the model's generalizability in detecting cyberbullying in visual content.
Certifications
Create Your First Automation Script Using Selenium and Java
Coursera
Supervised Machine Learning: Regression and Classification
DeepLearning.AI
Unsupervised Learning, Recommenders, Reinforcement Learning
DeepLearning.AI
Have a project
in mind?
Let's build it together.
Open to discussing AI/ML projects, generative AI products, full-stack engineering roles, training engagements, or just trading ideas about emerging technology.
Send a message