Passionate about AI/ML and modern web technologies, with hands-on experience building end-to-end solutions. Dedicated to delivering innovative applications and collaborating with experienced teams to solve real-world problems.
End-to-end
Projects
IEEE
Publication
Technical
Blogs
AI/ML Tools
Mastered
Python SQL REST APIs (FastAPI/Flask) Pydantic Supabase
LangChain Multi-Agent Systems LLM Routing RAG (Retrieval-Augmented Generation) Prompt Engineering
Llama 3 (Groq) OpenAI API Google Gemini IBM WatsonX Hugging Face TensorFlow Scikit-learn
FAISS Supabase (pgvector) Pandas NumPy Power BI Data Visualization
n8n (Automation) Google Cloud Platform (GCP) GitHub Streamlit VS Code Data Visualization
A proactive and passionate AI/ML developer with a proven track record of building end-to-end
solutions. My expertise in Retrieval-Augmented Generation (RAG) systems and generative models is
backed by hands-on experience developing tools like an autonomous web agent and a RAG-based
YouTube Q&A system. I've contributed to the field through an IEEE publication and several
technical blogs that simplify modern AI concepts. Committed to continuous learning, I'm eager to
apply my skills toward building impactful, real-world applications.
My professional goal is to contribute to impactful AI applications that bridge research with
real-world value, while continuously growing into roles that allow me to innovate at the
intersection of AI and practical problem-solving.
Outside of my technical work, I enjoy writing blogs to simplify complex AI concepts,
engaging in discussions on emerging technologies, and staying curious through continuous
learning.
Developed an intelligent request gateway that dynamically routes queries to optimal model tiers (Llama-3.1 8B/70B) based on real-time complexity scoring. Implemented a rule-based analyzer for reasoning depth and ambiguity detection to optimize inference efficiency.
Multi-agent Telegram chatbot with automated document ingestion, vector-based retrieval, and dual-model evaluation system. Features self-correction mechanism, web content analysis, and session memory for hallucination-free responses.
Developed an end-to-end machine learning pipeline to predict the landing success of SpaceX Falcon 9 first-stage rockets using historical launch data. Implemented data collection, EDA, feature engineering, model tuning, and interactive dashboarding to visualize and interpret classification results.
Built an end-to-end RAG system integrating YouTube Transcript API, LangChain, IBM WatsonX, FAISS, and embeddings to process video transcripts into searchable knowledge chunks for automated summarization and conversational Q&A.
Developed the AMGR framework integrating multi-modal data (text, image, video) with generative models to achieve enhanced recommendation accuracy and user engagement. Implemented reinforcement learning algorithms for real-time adaptation to dynamic user preferences.
View PublicationA beginner-friendly guide explaining the fundamentals of RAG systems and LLMs, making complex AI concepts accessible to developers and enthusiasts starting their journey in AI.
Read ArticleComprehensive guide on addressing one of the most critical challenges in modern AI systems - reducing hallucinations in RAG implementations through advanced techniques and best practices.
Read ArticleComplete implementation guide covering vector databases and semantic search technologies, providing practical insights into building efficient search systems for modern AI applications.
Read ArticleAvailable for remote work worldwide