I am a Gold Medalist and final-year Computer Science student at FAST University, Islamabad.
My core focus lies in Generative AI, Multi-Agent Systems, Computer Vision, and High-Performance Mobile Development.
I have hands-on experience in full-cycle AI/ML development β from data preprocessing and model training to evaluation and scalable deployment using modern MLOps practices.
Currently, I am expanding my expertise in AWS-native AI infrastructure, Kubernetes, and production-grade ML systems to build scalable and efficient AI applications.
- Designing advanced Multi-Agent RAG frameworks using GPT-4o, Gemini, LangChain, and LangGraph
- Building intelligent tutoring systems and virtual classroom architectures
- Working with Transformers including BERT, FinBERT, GPT-4o, and Gemini
- Developing deep learning pipelines using:
- YOLOv8
- U-Net
- CNN Architectures
- CycleGAN
- Working on semantic segmentation and real-time object detection systems
- Prompt Engineering (CoT, Few-shot Prompting)
- Multimodal Retrieval Systems
- RAG Pipelines with FAISS
- LLM Evaluation using:
- BERTScore
- BLEU
- ROUGE-L
- Building scalable mobile applications using:
- Flutter
- Kotlin
- Firebase
- SQLite
- MVVM Architecture
- Developing offline-first and real-time applications
- Learning scalable deployment workflows using:
- AWS (SageMaker, EC2, EKS, S3)
- Docker
- Kubernetes
- GitHub Actions CI/CD
- MLflow
- DVC
- Python
- Dart
- Kotlin
- Java
- C/C++
- C#
- SQL
- Assembly Language
- PyTorch
- TensorFlow
- scikit-learn
- Hugging Face
- LangChain
- LangGraph
- OpenAI APIs
- DDPG
- Flutter
- Android Development
- MVVM Architecture
- Firebase
- SQLite
- GetX
- FAISS Vector DB
- Redis
- Pandas
- NumPy
- Docker
- Git
- Streamlit
- Flask
- Jira
- Confluence
- Developed a virtual classroom utilizing GPT-4o/Gemini and LangGraph
- Orchestrated multi-agent dialogues between teachers and students
- Implemented a FAISS-based RAG pipeline
- Evaluated performance using:
- BERTScore
- BLEU
- ROUGE-L
- Built a processing pipeline for text and visual data from financial PDFs
- Utilized:
- CLIP for image embeddings
- Sentence-BERT for textual understanding
- Developed an interactive Streamlit interface
- Enhanced reasoning with Chain-of-Thought (CoT) prompting
- Combined YOLOv8 and a custom U-Net
- Performed recyclable waste detection and pixel-wise segmentation
- Applied advanced augmentations:
- Mosaic augmentation
- Color jitter
- Optimized performance metrics including:
- mAP
- IoU
- Developed a real-time Android social platform
- Integrated:
- Firebase Realtime Database
- Agora SDK
- Features include:
- Voice/video calling
- Vanish mode
- 24-hour stories
- Presence tracking
- Implemented scalable MVVM architecture
π Dec 2025 β Apr 2026
π Aug 2025 β Nov 2025
π Jul 2025 β Aug 2025
π May 2025 β Jul 2025
- Multi-Agent AI Systems
- Retrieval-Augmented Generation (RAG)
- Computer Vision Research
- AWS-native AI Deployments
- MLOps & Scalable Infrastructure
- Production-ready AI Applications
- πΌ LinkedIn: linkedin.com/in/muneeb-amir-6b3a9327b
- π» GitHub: github.com/muneeb-amir
- π§ Email: muneebamir24@gmail.com
β Passionate about building intelligent systems that combine AI research with scalable engineering.