Work with a variety of LLMs including Hugging Face OSS models, GPT (OpenAI), Gemini (Google), Claude (Anthropic), Mixtral (Mistral), and LLaMA (Meta).
Fine-tune and deploy LLMs for various use cases such as summarization, Q&A, RAG (Retrieval Augmented Generation), chatbots, document intelligence, etc.
Evaluate and compare model performance and apply optimization & MLOps :
Design and implement complete LLMOps workflows using tools like : MLFlow for experiment tracking and model versioning.
LangChain, LangGraph, LangFlow for LLM orchestration.
Langfuse, LlamaIndex for observability and indexing.
AWS SageMaker, Bedrock and Azure AI for model deployment and management.
Monitor, log, and optimize inference latency and model behavior in & Vector Stores :
Work with structured and unstructured data using MongoDB and PostgreSQL.
Leverage vector databases like Pinecone and ChromaDB for RAG-based applications.
Develop scalable data ingestion and transformation pipelines for AI training and & DevOps :
Deploy and manage AI workloads on AWS and Azure cloud environments.
Use Docker and Kubernetes for containerization and orchestration of LLM-based & Integration :
Build robust APIs and microservices using Python, with integrations using SQL and JavaScript where needed.
Develop UI interfaces or dashboards to visualize model outputs and system Skills :
Hands-on experience with multiple LLMs including GPT, Claude, Mixtral, Llama, etc.