About Me

AI Engineer and Python developer with hands-on experience building secure, scalable, cost, token and high-performance AI applications responsible. Specialized in GenAI systems, multi-agent workflows, RAG architectures, document intelligence, LLM’s, Model training and AI-powered automation. At BLDX and Commdel Consulting, Skilled in preparing high-quality datasets, designing data pipelines, training ML/DL models, and finetuning for real-world applications. I have led end-to-end development of AI components including document extraction systems, voice agents, multi-agent orchestration, and privacy preserving LLM pipelines.

Skills

Experience

Education

  • Programming Languages
    Python Java Bash
  • Machine Learning (ML)
    Sklearn OpenCV Pandas NumPy
  • Deep Learning (DL)
    TensorFlow Keras PyTorch
  • Generative AI
    huggingface Gemini Meta Ollama Mistral Claude AWS Bedrock
    Vector Embeddings
  • AI Agents & Orchestration
    langchain langgraph llamaindex LiveKit vapi
    RAG Pipelines
  • Web Frameworks
    Django Gradio FastAPI
  • Tools / Platforms
    huggingface Kaggle colab github git docker
    aws ECS, ECR, Bedrock
  • Databases
    MSQL
    MongoDB Atlas (Vector Store)
  • Junior Associate Engineer – AI
    Commdel Consulting Services Pvt. Ltd | Aug 2025 – Present
    Spearheaded the development of an AI-powered application with a Django backend, delivering features such as document named-entity extraction, intelligent document filtering, and a knowledge-aware voice agent integrated using LiveKit. Built an email assistant agent that automatically analyzes incoming user requests and generates context-aware, use-case-specific replies. Leveraged Gemini and ChatGPT to design and optimize AI workflows, significantly improving development speed and overall system productivity.
  • AI Engineer Intern (LLM & RAG)
    BLDX | Dec 2024 – Aug 2025
    Led AI development for a core project, ensuring secure data processing and strict client data privacy by preventing LLM training on sensitive information. Engineered a high-accuracy document search agent using LangGraph, capable of answering queries from 100+ complex document images. Implemented structured output formatting and integrated the system with a FastAPI backend, containerized with Docker and deployed on AWS.
  • LLM Systems & Multi-Agent Development
    Designed and deployed a Retrieval-Augmented Generation (RAG) agent for MongoDB-backed data using LlamaIndex for efficient retrieval and LangChain for LangGraph integration. Conducted in-depth data analysis on client project data to identify risks, warnings, and actionable insights using advanced prompt engineering techniques. Currently developing a scalable multi-agent AI system by orchestrating multiple specialized agents to enhance overall system intelligence and performance.
  • Bachelor of Technology (B.Tech)
    Computer Science and Engineering - Specialization in AI & Machine Learning
    Institution: Vasireddy Venkatadri Institute of Technology, Guntur
    CGPA: 8.62
    Duration: 2021 - 2025
    Graduated: Yes
  • Class XII
    Science Stream
    Institution: Narayana Junior College, Andhra Pradesh Board of Intermediate Education
    Percentage: 92.2%
    Duration: 2020 - 2021
  • Class X
    General Education
    Institution: Narayana English Medium School, Andhra Pradesh Board of Secondary Education
    CGPA: 9.7/10 (92.15%)
    Duration: 2018 - 2019

certification

Programming


    Python

  • Kaggle
  • SoloLearn
  • HackerRank
  • GUVI

    Java

  • NPTEL (ELITE)

AI&ML


    Machine Learning

  • Coursera
  • Kaggle
  • Microsoft
  • IBM

    Deep Learning

  • Kaggle
  • NPTEL
  • OpenCVUniversity
  • IBM

Miscellaneous


    Google Skill Boost

  • Profile

    Data Science

  • NPTEL (ELITE+🥈)

    Django

  • Microsoft
  • Udemy

Latest Projects

DearHRSpeakWithMy

DearHRSpeakWithMy 2.0

Python LLaMA3-8B Gradio HuggingFace

DESCRIPTION: AI-powered HR Interview Assistant using LLaMA3 on Hugging Face. Generates structured, professional answers based on candidate profile, optimized for mock interviews.


  • Integrated prompt-engineered answers with real-time feedback and profile enrichment.
  • Deployed on Hugging Face Spaces with an interactive chatbot interface.
Document RAG Agent

Document Query AI Agent

Python LlamaIndex LangGraph Gradio MongoDB

DESCRIPTION: Developed a high-accuracy document Q&A system using RAG and LLMs. Vector embeddings + LlamaIndex + MongoDB handled large image/document-based queries.


  • Utilized semantic embedding + query retrial mechanism to boost answer relevance.
  • Integrated with structured LangGraph agent pipelines and hosted via Gradio.
AI Agent Nexus

AI Agent Nexus

Python CrewAI Gemini Gradio HuggingFace

DESCRIPTION: Multi-agent AI suite using CrewAI and Gemini. Agents perform SEO writing, game dev automation, and marketing content creation.


  • Built structured workflows for agent collaboration and task handoff.
  • Fully modular; supports dynamic task injection and role-based prompt strategies.
Epic Minds

Epic Minds (RamayanaGPT & GitaGPT)

Python LlamaIndex Gemini Gradio MongoDB HuggingFace

DESCRIPTION: RAG-based QA system for Indian scriptures with multilingual embeddings and Gemini Flash LLM. Supports spiritual exploration and educational queries.


  • Embedded multilingual scriptures and created vector search index.
  • Deployed with Gradio + collapsible UI components + real-time logs.
PC AI Assistant

AI Assistant for PC

Python Langchain MCP Gemini Gradio

DESCRIPTION: Developed an agent-based desktop assistant with modular servers for file, info, and search tasks using LangChain. Designed for MPC-style automation.


  • Custom-designed file_server, info_server, and web_search_server modules.
  • Designed to handle contextual tasks and future voice control integration.
VGGFace

Face Recognition with VGGFace

Python TensorFlow Keras OpenCV Hugging

DESCRIPTION: Built a 31-class face recognition app with 97.86% accuracy using VGGFace + TensorFlow. Hosted via Django + Gradio.


  • Mastered OpenCV-based dataset creation, and trained CNNs with high generalization.

SCRAM

python TensorFlow Keras Gemini OpenCV Django

DESCRIPTION: Engineered a robust, secure Django-based system for comprehensive college resource and access management, incorporating advanced features such as user management, complaint registration, and a chatbot assistant. Seamlessly integrated Gemini-Pro LLM with Google API and introduced face recognition technology for enhanced attendance tracking.


  • Devised robust user authentication and 2-FS password authentication, enhancing system security and reliability.
  • Led the project team, developing comprehensive Django templates and seamlessly integrating custom chatbot functionalities using LLM.

Object Segmentation

OpenCV YoloV8 TensorFlow Keras OpenCV

DESCRIPTION: Developed a versatile Python script leveraging OpenCV and scikit-image for advanced image processing tasks, including object segmentation using YOLOv8. Implemented functionalities such as grayscale contrast enhancement, image similarity measurement (SSIM), and object detection (YOLO).


  • Engineered robust image processing capabilities with detailed evaluation and visualization of results, demonstrating significant improvements in processing accuracy and efficiency.
  • Planned future integration of advanced deep learning models to expand functionality and further enhance system performance.

MNIST: CNN Architectures Comparison

Python TensorFlow Keras OpenCV

DESCRIPTION: Compared deep learning architectures like LeNet, AlexNet, ZFNet and VGG16 for digit recognition, achieving 99.55% accuracy using transfer learning. Applied CNNs effectively for image recognition tasks. Optimized training with data preprocessing and learning rate scheduling.


  • Achieved high accuracy with AlexNet and LeNet architectures on the MNIST dataset.
  • Utilized transfer learning to enhance model performance.
  • Applied data preprocessing and learning rate scheduling techniques to optimize training.
Repository Image

Academic Labs and Projects Repository

Django python Git

DESCRIPTION: This repository contains academic lab work, additional learnings, and basic projects. Created during my second year of B-Tech, it showcases various resources to help students learn and practice skills in subjects like programming, Django, and Machine Learning. I'm grateful to my HOD for introducing me to version control systems and GitHub.


  • Provides resources for students to enhance their skills in programming, Django, and Machine Learning.
  • Includes examples of simple projects for students to practice and improve their understanding of various subjects.
See more

Contact Me

bandreddysarathchandra@gmail.com

+91 6305419947

Download Resume

CopyRight ©️ Sarath