🎓 Undergraduate at SLIIT – BSc (Hons) in Information Technology 🧠 Passionate about AI/ML, Generative AI (GenAI), MLOps, and building real-world applications
I'm currently engaged in two major projects, leveraging my passion for MLOps and Generative AI to build robust, real-world applications:
- Developing a robust Retail Demand Forecasting System for product demand prediction.
- Model Benchmarking: Comparing classical time series models (Prophet, SARIMA) with advanced deep learning architectures (Temporal Fusion Transformer - TFT).
- Advanced Techniques: Focused on creating hybrid model ensembles to maximize forecasting accuracy.
- Deployment & Visualization: Building an interactive user interface using Streamlit to visualize forecasts and key business insights.
- Designing and implementing a conversational Tax Supporter system to provide guidance and answer user queries related to tax filing and regulations.
- Agentic Framework: Utilizing LangGraph to build a complex, multi-agent architecture (e.g., dedicated agents for Document Retrieval, Regulation Checking, and Conversational Flow).
- Core Technology: Focusing on defining clear state transitions and orchestrating the agents to handle complex, multi-step queries accurately and efficiently.
- ⚡ Implementing MLOps pipelines for efficient model deployment and monitoring
- 🧠 Deep dive into Transformer Architecture and the mechanics behind large-scale LLMs
- 🤖 Deep dive into Large Language Models (LLMs) and Generative AI applications
- 🌐 Building Agentic AI systems and robust RAG (Retrieval-Augmented Generation) workflows
- ☁️ Mastering AWS Cloud Services (SageMaker, Lambda, S3) for scalable AI solutions
- 🚀 Exploring Go (Golang) for building high-performance, concurrent backend services, especially for MLOps deployment
- 📧 Email: chanupadeshan2002@gmail.com
- 💼 LinkedIn: chanupadeshanmunasinghe
- 💻 GitHub: chanupadeshan