Vamsidhar Vuddagiri

About

AI/ML engineer who led Impacter AI’s production deployment of a specialized outreach LLM (CharmedLLM) fine-tuned on GPT-4.1, cutting API costs ~40% while boosting outreach effectiveness ~60%. Built the supporting MLOps and data infrastructure (MLflow, Kubernetes, PySpark, Kafka) and has agentic AI experience from University of Dayton, using LangChain + RAG and vector search (Pinecone) to improve reliability and reduce hallucinations.

Experience

Education

Key Strengths

• Led development and deployment of a domain-specific LLM for personalized sales outreach (CharmedLLM)
• Reduced LLM/API costs by ~40% through model strategy and optimization
• Improved outreach effectiveness by ~60% with fine-tuned model and workflow changes
• End-to-end MLOps: model validation/versioning/iteration with MLflow and container orchestration with Kubernetes
• Built automated high-volume data pipelines with PySpark to convert unstructured data into fine-tuning-ready formats
• Applied SFT then transitioned to RLHF to maintain performance and adapt to new data
• Designed reliable agentic workflows using LangChain + RAG to reduce hallucinations and improve grounding
• Strong model/retrieval/prompting selection judgment (cost/complexity tradeoffs; dense retrieval with Pinecone; ReAct/multi-chain prompting)
• Effective cross-functional communication with sales stakeholders using plain language and visual comparisons (graphs/charts)