I'm an AI/ML Engineer focused on building practical, production-oriented AI applications using Generative AI, Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), multimodal AI, and Python backend engineering.
I enjoy turning complex problems into reliable AI-powered systems, from intelligent document processing and information retrieval to API-driven AI workflows and scalable backend services.
- 🔭 Building: LLM-powered applications, document intelligence, AI automation, and intelligent workflows.
- 🧠 Exploring: Agentic AI, multimodal reasoning, LLM fine-tuning, and scalable inference.
- 🤖 AI Focus: RAG pipelines, LLM orchestration, embeddings, vector search, OCR, and vision-language models.
- ⚙️ Backend: Python, FastAPI, REST APIs, asynchronous processing, Celery, and Redis.
- ☁️ Cloud: AWS, Azure, and Google Cloud technologies.
- 🌱 Continuous Learning: Model optimization, inference performance, AI system design, and MLOps.
- 💡 Interests: Applied AI, Generative AI, NLP, computer vision, and AI product engineering.
Core concepts: Prompt Engineering, Embeddings, Semantic Search, Retrieval Pipelines, Context Management, Structured Outputs, Tool Calling, and LLM Evaluation.
Areas of interest: NLP, Computer Vision, Document AI, OCR Pipelines, Vision-Language Models, Model Fine-Tuning, and Inference Optimization.
Core concepts: Vector Indexing, Similarity Search, Metadata Filtering, Retrieval Optimization, and Database Design.
Core concepts: API Design, Async Processing, Background Jobs, Service Integration, Error Handling, and Production Backend Systems.
An AI-powered support assistant designed around knowledge retrieval and intelligent response generation.
- Retrieval-Augmented Generation (RAG)
- Semantic Search and Vector Databases
- LLM-Based Response Generation
- Workflow Orchestration and Tool Integration
Tech: Python, FastAPI, LLMs, LangChain, Vector Search
An intelligent document-processing workflow combining document images, OCR, vision-language models, structured extraction, and validation rules.
- Document Understanding and Information Extraction
- OCR and Vision-Language Model Integration
- Structured Data Extraction
- Rule-Based Validation and Verification
Tech: Python, OCR, PyTorch, VLMs, FastAPI
A knowledge retrieval application designed to retrieve relevant context from documents and use it to generate grounded responses.
- Document Ingestion and Chunking
- Embedding Generation
- Vector Similarity Search
- Context Retrieval and LLM Integration
Tech: Python, LangChain, FAISS, MongoDB, LLMs
A backend architecture for integrating AI capabilities into application workflows through APIs and asynchronous processing.
- REST API Development
- Asynchronous Task Processing
- AI Service Integration
- Database and Cache Integration
Tech: Python, FastAPI, Celery, Redis, SQL
Note: The project descriptions above are portfolio concepts. Replace the links with the corresponding repositories and adjust each description to match the code you have actually published.
- Building reliable Generative AI applications.
- Improving RAG quality, retrieval accuracy, and response grounding.
- Exploring agentic workflows and tool-using AI systems.
- Fine-tuning open-source models for specialized tasks.
- Developing scalable, maintainable Python backend services.
- Learning more about inference optimization, deployment, and AI system design.
"Building AI systems that are useful, reliable, and impactful."