Crafting intelligent systems at the intersection of Agentic AI, Context Engineering, and scalable ML infrastructure.
I design AI that plans, retrieves, adapts, and acts with real autonomy.

• 🧠 Agentic AI — tool-use graphs, planning loops, autonomous workflows
• 🧩 Context Engineering — retrieval orchestration, memory layers, long-context optimization
• 🎥 Multimodal Intelligence — vision–language models, audio/text fusion, streaming inference
• 🚀 AI Core Tech — transformers, embeddings, RL, diffusion models
• 🖥️ Infra & Systems — distributed training, GPU orchestration, scalable model serving
• 🗂️ Retrieval Systems — vector search, hybrid retrieval, semantic indexing, graph-enhanced RAG

• High-throughput inference pipelines
• Intelligent, tool-using AI agents
• Retrieval systems with adaptive context routing
• Cloud-native ML platforms (AWS · GCP · Azure)
• Real-time AI applications using Kafka + Spark

Agentic AI · Context Engineering · Multimodal Models · RAG Systems · Distributed Training · Vector Databases · MLOps Architecture

AI/ML: PyTorch · HuggingFace · JAX
Agentic/RAG: LangGraph · LlamaIndex · FAISS · Milvus · Elastic
Infra: Kubernetes · Docker · Terraform · Ray · MLflow
Data: Kafka · Spark · dbt · Snowflake
Cloud: AWS · GCP · Azure

Cut LLM latency and cost dramatically by restructuring the retrieval flow and context window strategy — twice the performance, zero extra model size.