I build production-grade AI solutions leveraging Large Language Models, Agentic AI, and Retrieval-Augmented Generation systems. With expertise in LLM integration, multimodal AI, and scalable vector database architectures, I transform complex business challenges into intelligent, autonomous applications. Currently driving AI innovation in healthcare at Connected Value Health.
02+
15+
03+
Currently working at Connected Value Health, I build production-grade AI solutions for healthcare using FastAPI, LLMs, and Vector Databases. I specialize in Agentic AI, Retrieval-Augmented Generation (RAG), Multimodal AI, and LLM fine-tuning. With expertise in LangChain, LlamaIndex, Gemini, and model optimization, I design autonomous AI agents and intelligent applications that solve complex real-world problems. Passionate about pushing the boundaries of Generative AI.
Know MoreWorked on real-time healthcare project building AI-powered solutions using FastAPI, LLMs, and Mistral OCR for automated medical document processing and information extraction.
Developed production-grade RAG systems with Vector Databases (Milvus, Pinecone) for intelligent healthcare data retrieval and personalized patient information delivery.
Sep 2025 - PresentDeveloped MediSaga, a RAG Chatbot Application for medical question answering using RAPTOR indexing with Milvus vector database and Llama2.
Built a High-Performance CPU RAG System leveraging Quantized Llama2 to optimize retrieval-augmented generation on CPU.
Demonstrated expertise in natural language processing, model optimization, and vector database integration.
Jul 2024 - Aug 2024 View CertificateDeveloped a LangChain-based Medical Assistant app utilizing OpenAI GPT-3.5 model for accurate medical information retrieval.
Designed and implemented a Checkbox Detection app using OpenCV to enhance data processing for Vizhi EyeCare NGO.
Demonstrated proficiency in machine learning, natural language processing, and computer vision, contributing to impactful projects.
Dec 2023 - Jan 2024 View CertificateI'm currently a Data Science student, I have a strong foundation in algorithms, data structures, software design and Development. My coursework has provided me with a deep understanding of the principles and practices of Computer Science, Data Science , Machine learning, Artificial Intelligence and I have honed my programming skills through various projects and assignments.
2022- presentAs computer science student, i had learned the basic concepts of computer,its working and programming language, where I'm gained hands-on experience in developing and designing software solutions. Such training provides me with a solid foundation in software development , learning the fundamentals of software development and building projects using the Full Stack.
2018 - 2021
KPR Institute of Engineering and Technology
Skills Acquired: Computer Vision workflows, NLP pipelines, Generative AI prototyping, data analytics, and real-time AI application development.
Show Credential
Skills Acquired: Prompt engineering, Gemini usage in Vertex AI, multimodal prompting, image analysis, and response quality tuning.
Show Credential
Skills Acquired: MLOps lifecycle, CI/CD for ML, model deployment, training and inference automation, and cloud-based ML operations.
Show Credential
Skills Acquired: Transformer architecture, self-attention, BERT fundamentals, text classification, question answering, and NLP task design.
Show Credential
Sector Skill Council Nasscom
Skills Acquired: Data categories and metadata, data collection methods, data storage basics, pandas DataFrame handling, and data validation techniques.
Show Credential
Coursera
Skills Acquired: Exploratory data analysis, statistical visualization, dimensionality reduction, cluster-based pattern discovery, and analytical storytelling.
Show Credential
Skills Acquired: Analytical thinking, data lifecycle understanding, spreadsheet and visualization tool usage, data ethics, and foundational analytics practice.
Show Credential
Coursera
Skills Acquired: NumPy array operations, pandas DataFrame manipulation, indexing and filtering, data wrangling, and tabular analysis in Python.
Show Credential365 Data Science
Skills Acquired: Git fundamentals, branching workflow, commit history tracking, diff and log analysis, and GitHub collaboration practices.
Show Credential
IBM
Skills Acquired: Supervised and unsupervised learning, regression and classification, model evaluation, feature engineering, and scikit-learn implementation.
Show Credential
TruthLens is an AI-powered fake news detection system built with a fine-tuned BERT model. The project features a FastAPI backend, React/Tailwind frontend, MongoDB authentication, and real-time news verification using Google Search API. This comprehensive solution enables users to verify the authenticity of news articles instantly, combating misinformation with state-of-the-art NLP techniques. The system provides confidence scores and source verification to help users make informed decisions about the content they consume.
View Project
GARUDA is a high-precision surveillance system that leverages the YOLOv8 AI model to detect people in real-time video from drones or other cameras. The system includes a web control panel and instant Telegram notifications with encrypted communications for enhanced security. Built with Flask backend, it features computer vision capabilities for real-time object detection, making it ideal for security monitoring, crowd analysis, and automated surveillance applications.
View Project
MediSaga is an intelligent medical question-answering system that combines Retrieval-Augmented Generation (RAG) with advanced indexing techniques using RAPTOR and Milvus vector database. Powered by Llama2, it delivers accurate and contextually relevant answers to medical inquiries. The project features robust PDF text extraction, query expansion capabilities, and a seamless Streamlit-based UI for healthcare professionals and individuals seeking precise medical information.
View Project
SkyQuery is an AI-powered flight information assistant using Flask, MongoDB, and Llama 3.3 70B. It supports natural language queries, user role management, and an admin dashboard for flight data management. The system leverages agentic AI and conversational interfaces to provide seamless access to flight information, making it easy for users to query complex database information using simple natural language.
View Project
An intelligent FAQ Chatbot that answers questions with contextual understanding using FastAPI, LangChain, and ChromaDB. The system maintains conversation context across multiple turns, processes and indexes documents dynamically, and provides accurate responses based on the knowledge base. This project demonstrates expertise in building production-ready conversational AI systems with advanced memory and retrieval capabilities.
View Project
An Enterprise Expense Automation System built using Model Context Protocol (MCP) and NLP. This intelligent expense tracking solution automates expense categorization, receipt processing, and financial reporting using natural language processing. The system features a client-server architecture with MCP integration, enabling seamless communication between AI models and expense management tools for automated data extraction and smart categorization.
View Project
AUTOMATOS is an advanced automated surveillance and messaging system designed to enhance security and monitoring capabilities. Built with Python and state-of-the-art YOLO models (YOLOv5 and YOLOv8) for real-time detection and identification of individuals and activities. The system provides automated alerts and integrates with messaging platforms, making it ideal for organizations seeking to bolster their security infrastructure.
View Project
Text2SQL is an intelligent system that converts natural language queries into SQL statements. Leveraging advanced NLP and LLM capabilities, this project enables users to interact with databases using plain English, eliminating the need for SQL expertise. Perfect for business analysts and non-technical users who need to extract insights from databases without writing complex queries.
View Project
A Multimodal Retrieval-Augmented Generation system that processes and understands both text and images. This advanced RAG implementation combines visual and textual data to provide comprehensive answers, enabling users to query documents containing mixed media. Built with cutting-edge vision-language models for enhanced document understanding and information retrieval.
View Project
"Surya has been an exceptional ML engineer in our team. His ability to swiftly grasp complex machine learning concepts and implement them in real-world applications is remarkable. During his time with us, he developed a highly accurate Medical Assistant app that streamlined information retrieval, benefiting our healthcare clients significantly. Surya's dedication to quality and innovation is evident in every project he undertakes. His expertise in NLP, computer vision, and efficient model deployment makes him a valuable asset to any team."
"Surya’s contributions as an ML engineer have been instrumental in advancing our AI-driven projects. His work on the 'MediSaga-RAG-Chatbot-Application' showcased his deep understanding of state-of-the-art ML techniques, particularly in retrieval-augmented generation and vector database integration. Surya not only excels technically but also possesses a keen ability to collaborate across teams, ensuring that projects are delivered on time and exceed expectations. His passion for machine learning and his continuous pursuit of knowledge make him a standout professional."
"It has been a pleasure working with Surya, who is an outstanding ML engineer. His innovative approach to problem-solving and his expertise in deploying scalable machine learning solutions have consistently impressed our team. Surya played a pivotal role in developing and optimizing our ML models, resulting in significant performance improvements. His commitment to staying ahead of industry trends and his proactive attitude towards challenges make him a reliable and forward-thinking engineer."
Ready to work together on your next software project? I'd love to hear from you! Feel free to get in touch using the contact information below. Whether you have a specific project in mind or just want to chat about your software development needs, I am here to help. I will respond to your message as quickly as possible and look forward to connecting with you soon.