Vidyasagar Karuturi

Vidyasagar Karuturi

GenAI Engineer & Data Scientist | Crafting Intelligent Solutions

6309807183 karuturividyasagar@gmail.com LinkedIn Medium GitHub

Professional Summary

GenAI Engineer and Data Scientist with a proven track record in designing, developing, and deploying end-to-end AI/ML solutions at scale. Deep expertise in Large Language Models. Skilled in model training, fine-tuning, and evaluation, with proficiency in deep learning frameworks like TensorFlow and PyTorch, and strong command of MLOps principles. Demonstrated success in building scalable machine learning pipelines, deploying production-grade AI systems and leveraging cloud platform like AWS for real-time AI applications. Seeking a challenging and rewarding role at a leading technology company where I can contribute to innovative AI products and advancements.

Core Technical Skills

Programming Languages & Frameworks

  • Python (NumPy, Pandas, Scikit-learn, Keras, Matplotlib)
  • SQL

Databases

  • MongoDB
  • MySQL

Machine Learning

  • Supervised Learning (Decision Trees, Random Forests)
  • Unsupervised Learning
  • Data Modeling & Evaluation
  • Preprocessing & Postprocessing
  • Model Optimization & Performance Tuning

Rest-API

  • FastAPI
  • Flask

Deep Learning

  • PyTorch
  • TensorFlow

Natural Language Processing

  • Classification, Summarization, Text Generation, Name Entity Recognition
  • Transformers, Retrieval Augmented Generation, ReAct, Retrieval, AgenticRAG
  • Langchain, Llamaindex, FastMCP, Ollama, Groq, PydanticAI

LLM Experience

  • Meta LLaMA 2, Google Gemma, OpenAI GPT-4, Anthropic Claude
  • MCP, CrewAI, Google ADK, Google VertexAI

Cloud & MLOps

  • Amazon Web Services (EC2, S3, IAM, Lambda, RDS, Sagemaker)
  • Azure, Terraform, Docker, Kubernetes
  • CI/CD via Jenkins, Git, Postman

Tools

  • Visual Studio Code, Cursor, Jupyter Notebook, CVAT
  • PuTTY, Tableau, Langfuse, MLflow, Kibana

Professional Experience

Data Scientist Intellect Design Arena Pvt. Ltd., Chennai

Jun 2023 – Present

  • Designed and scaled a complex GenAI platform for leading financial firms to process, extract, and classify millions of diverse financial documents, with real-time data streams and scalable knowledge bases, achieving latency under 200ms, 97.9% uptime, and seamless horizontal scaling.
  • Developed and integrated an MVP on Reasoning and Action Agents, Chain-of-Thought (CoT) and Tree-of-Thought (ToT) using Langchain acquisition, feature engineering, model development, evaluation, and cloud deployment using PyTorch and TensorFlow.
  • Built LLM-based AI Guardrails for Prompt Injection Detection, Personally Identifiable Information (PII) masking, and Toxic Content Filtering using Meta LLaMA 2 70B, GPT-4o, and Gemma 2B with SETFIT and AWS Comprehend which safeguards 80% of attempts.
  • Integrated Smart Retrieval-Augmented Generation (RAG) with short term and long term memory using Redis and Cosmos DB to improve contextual awareness and user experience in financial service dialogues, while also implementing Cache Augmented Generation (CAG) to enhance response speed and reduce latency by preloading knowledge into the model’s context window.
  • Designed and implemented an automated evaluation module to assess LLM and LMM-generated responses across Correctness, Faithfulness, and Relevancy metrics, leveraging GPT-4 and GPT-4o-based evaluators (LLM-as-aJudge and ROUGE-L, fuzzy JSON, exact match scoring, and accuracy metrics) for comprehensive quality assessment in RAG workflows.

Cloud Intern LTI Mindtree

Feb 2023 – Apr 2023

  • Automated Infrastructure as Code (IaC) solutions using Terraform for AWS resource provisioning (EC2, S3, IAM, Lambda, RDS).
  • Implemented security compliance and cost optimization strategies for cloud-based deployments.

Education

Bachelors of Technology in Computer Science and Engineering (Specialization: Data Analytics)

Vellore Institute of Technology University, Amaravati

2019 – 2023 | CGPA: 8.05 / 10

Relevant Coursework: Linear algebra, Object Oriented Programming, Databases, Discrete Mathematics, statistics, Operating Systems, Computer Networks, Machine Learning, Data Mining, Cloud computing

Intermediate Education (Science Stream)

Narayana Junior College, Vijayawada

2017 – 2019 | CGPA: 9.5 / 10

Secondary Education (10th Grade)

St. Claret EM High School

2017 | CGPA: 9.7 / 10

Projects

AI Guardrails System

Implemented robust guardrails using SETFIT and LLMs to detect prompt injection, PII exposure, and harmful content. Applied synthetic data generation via Gemma 2B for anonymization.

Citation Attribution Engine

Engineered citation backfilling to validate and attribute references used in generative responses, increasing more transparency and granularity.

LLM Evaluation Framework

Developed metrics-based evaluation pipeline for assessing benchmarking agent outputs using Judge, ROUGE-L, fuzzy matching, and JSON schema validation.

Automated Document Intelligence

Created a document classification and extraction model based on LayoutLM for automating categorization of emails and financial reports, improving efficiency by 94% on unstructured data.

Reach Out

Send an Email

Click the button below to open your email client and send me a message.

Send Email