- MSc Artificial Intelligence student @ Brandenburgische Technische Universität Cottbus-Senftenberg (October, 2027 (Expected))
- Former Software Engineer with 4+ years of experience @ Millennium Information Solution Ltd
- Experienced in building Enterprise Software Solutions using Spring Boot and Angular. Developed applied AI projects in NLP, RAG, AI Agents, LLMs and Computer Vision
- Learning and implementing core concepts of Machine Learning, Deep Learning, Generative AI and Computer Vision.
- Ask me about Python, AI/ML, NLP, Computer Vision, LLMs, MLOps, Cloud (AWS, Kubernetes)
- 2025 goals: Land an AI internship/research role and contribute to open-source LLM projects
Languages: Python, Java, TypeScript, PL/SQL
Frameworks & Libraries: PyTorch, TensorFlow, OpenCV, Scikit-learn, LangChain, LangGraph, FastAPI, Spring Boot, Angular
DevOps & MLOps: Docker, Kubernetes, DVC, MLflow, GitHub Actions (CI/CD), Apache Airflow
Cloud Platforms: AWS (EC2, S3, Lambda, SageMaker)
Databases: PostgreSQL, MongoDB, FAISS, Pinecone, ChromaDB
Claim-Polygraph extracts and verifies claims from articles, web pages, and YouTube videos using LLM and fact-checking APIs, with a web interface for analysis and review.
Tech Used: Python, OpenAI API, Google Fact Check Tool, ClaimBuster API
- Paste raw text, a news/article URL, or a YouTube video URL into the input box.
- The system will extract text, identify claims, and run fact-checking.
- Results include:
- Top claim-worthy sentences (ClaimBuster)
- LLM-based fact-checks with verdicts, confidence, reasoning, and sources (OpenAI GPT)
- External fact-check reviews from Google Fact Check Tools
Real-time German traffic sign detection and tracking using YOLO11 and ByteTrack. Supports live webcam/RTSP streams, video files, and images (with sequence tracking).
Tech Used: Python, OpenCV, PyTorch, YOLO11, NumPy, Pandas
- Detects German traffic signs from live video feed, video files and images
- Multi-object Tracking using ByteTrack (stable IDs across frames).
An AI agent powered by Retrieval-Augmented Generation (RAG) that scans the internet and internal knowledge bases to assist book buyers by providing relevant and accurate information
Tech Used: Python, React, BLIP Model, SerpAPI, LangChain, FAISS, OpenAI GPT-3.5-turbo
- Automates retrieval & reasoning for book-related queries
- Seamless integration with multiple knowledge sources
Doctor's Mind - a chatbot empowering doctors with intelligent decision-making assistance.
Tech Used: Python, React, LangChain, Hugging Face, Pinecone Vector DB, Mistral-7B-Instruct
- Provides intelligent query handling for medical knowledge
- Built with retrieval-augmented generation for accuracy
- MSc Artificial Intelligence — Brandenburgische Technische Universität Cottbus-Senftenberg, Germany (2023 – Present)
- BSc Computer Science and Engineering — American International University - Bangladesh (2013 – 2016)
- AI/ML, NLP and Computer Vision Research & Engineering Internships, Working Student or Full-Time Role
- Collaborations in LLMs, RAG, Applied NLP and Computer Vision
- Open-source contributions in AI, Cloud, and MLOps
- Email: moshiur.mishuk@gmail.com
- LinkedIn: moshiur00
- GitHub: moshiur00
Thanks for visiting my profile! Feel free to connect, collaborate, or just say hi