Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I'm Varshith N, an AI/ML Engineer with 4+ years of experience delivering end-to-end ML and Generative AI solutions across telecom, finance, and insurance. I specialize in Large Language Models, MLOps, computer vision, NLP, data engineering, and graph-based AI systems. I thrive on building production-grade pipelines for batch, streaming, and real-time inference, including agentic workflows, tool calling, and API-driven architectures. I enjoy designing retrieval-augmented generation systems, semantic search, and scalable deployments on AWS, Azure, and GCP, with a focus on data quality, governance, and responsible AI. I collaborate with cross-functional teams to automate decision-making, reduce manual effort, and deliver measurable business impact.…I'm Varshith N, an AI/ML Engineer with 4+ years of experience delivering end-to-end ML and Generative AI solutions across telecom, finance, and insurance. I specialize in Large Language Models, MLOps, computer vision, NLP, data engineering, and graph-based AI systems. I thrive on building production-grade pipelines for batch, streaming, and real-time inference, including agentic workflows, tool calling, and API-driven architectures. I enjoy designing retrieval-augmented generation systems, semantic search, and scalable deployments on AWS, Azure, and GCP, with a focus on data quality, governance, and responsible AI. I collaborate with cross-functional teams to automate decision-making, reduce manual effort, and deliver measurable business impact.

Varshith N

AI Engineer, Data Scientist, Full Stack Developer, +3





I'm Varshith N, an AI/ML Engineer with 4+ years of experience delivering end-to-end ML and Generative AI solutions across telecom, finance, and insurance. I specialize in Large Language Models, MLOps, computer vision, NLP, data engineering, and graph-based AI systems. I thrive on building production-grade pipelines for batch, streaming, and real-time inference, including agentic workflows, tool calling, and API-driven architectures. I enjoy designing retrieval-augmented generation systems, semantic search, and scalable deployments on AWS, Azure, and GCP, with a focus on data quality, governance, and responsible AI. I collaborate with cross-functional teams to automate decision-making, reduce manual effort, and deliver measurable business impact.…I'm Varshith N, an AI/ML Engineer with 4+ years of experience delivering end-to-end ML and Generative AI solutions across telecom, finance, and insurance. I specialize in Large Language Models, MLOps, computer vision, NLP, data engineering, and graph-based AI systems. I thrive on building production-grade pipelines for batch, streaming, and real-time inference, including agentic workflows, tool calling, and API-driven architectures. I enjoy designing retrieval-augmented generation systems, semantic search, and scalable deployments on AWS, Azure, and GCP, with a focus on data quality, governance, and responsible AI. I collaborate with cross-functional teams to automate decision-making, reduce manual effort, and deliver measurable business impact.

Available to hire

I’m Varshith N, an AI/ML Engineer with 4+ years of experience delivering end-to-end ML and Generative AI solutions across telecom, finance, and insurance. I specialize in Large Language Models, MLOps, computer vision, NLP, data engineering, and graph-based AI systems. I thrive on building production-grade pipelines for batch, streaming, and real-time inference, including agentic workflows, tool calling, and API-driven architectures.

I enjoy designing retrieval-augmented generation systems, semantic search, and scalable deployments on AWS, Azure, and GCP, with a focus on data quality, governance, and responsible AI. I collaborate with cross-functional teams to automate decision-making, reduce manual effort, and deliver measurable business impact.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Language

English

Fluent

Work Experience

Generative AI Engineer at Verizon Wireless Systems

May 1, 2024 - Present

Architected a multi-agent workflow using LangGraph, Neo4j, and Elasticsearch to automate feature generation from business requirements, reducing manual solution architect effort by 40% and accelerating SDLC pipeline efficiency. Orchestrated autonomous and human-in-the-loop agents leveraging LLMs and FastAPI, enabling iterative refinement of generated features through conversational feedback, which enhanced accuracy and contextual alignment by 25%. Integrated heterogeneous data sources into a unified knowledge graph (graph-based ontology) using Neo4j Database, enabling context-aware retrieval and RAG-style generation that accelerated insight delivery by 35%. Deployed custom tools on MCP servers, allowing LLMs to autonomously select optimal tools for feature generation and enabling external clients like Copilot to execute SDLC automation tasks 50% faster with improved interoperability across systems. Engineered a graph-driven migration pipeline that leveraged LLMs to translate legacy COB

Applied AI Engineer at Liberty Mutual Finance

July 1, 2023 - April 1, 2024

Designed a Retrieval-Augmented Generation (RAG) pipeline using LangChain, FAISS, and OpenAI GPT-4 for contextual document-grounded answers. Integrated FastAPI, Docker, and AWS Lambda to deploy low-latency LLM APIs for production inference. Developed custom embeddings for insurance FAQs and claims data using HuggingFace Transformers and Sentence-BERT. Created prompt templates with PII masking and policy validation to ensure secure and compliant conversations. Engineered a fallback intent classifier using SVM (scikit-learn) to divert high-risk inputs to human agents. Logged all model interactions in MongoDB and AWS S3, enabling auditable and retrainable inference pipelines. Tuned BLEU/ROUGE scores weekly via human-in-the-loop review to refine LLM response quality and structure. Achieved 42% faster responses and automated 60% Tier-1 queries, improving CSAT scores with over 90% accuracy. Applied guardrails using policy rules and tracking to mitigate hallucinations and ensure regulatory com

MLOps Engineer at Sundaram Finance

July 1, 2020 - June 1, 2022

Built MLflow-integrated pipelines to track model versions, parameters, datasets, and training-to-deployment lineage. Automated drift detection using Kolmogorov–Smirnov test and PSI, alerting teams of distribution shifts. Integrated Great Expectations with ETL pipelines to validate schema, detect nulls, and enforce data quality. Enforced CI/CD workflows using GitHub Actions and MLflow Registry for controlled model promotion processes. Visualized training and production metrics using Weights & Biases, Grafana, and custom alerting dashboards. Generated SHAP and LIME explainability outputs for each model prediction reviewed by legal and business teams. Implemented responsible AI checks, covering bias tests, fairness metrics, and data consent validation. Automated model documentation with ethical concerns, assumptions, input features, and validation stats included. Reduced model failures by 70% through real-time monitoring, explainability, and traceable ML practices. Passed 2 regulatory a