Skip to content
View liteshperumalla's full-sized avatar
🎯
Focusing
🎯
Focusing
2 followers · 2 following

Achievements

Achievement: Pull Sharkx2Achievement: Pair ExtraordinaireAchievement: YOLOAchievement: Quickdraw

Achievements

Achievement: Pull Sharkx2Achievement: Pair ExtraordinaireAchievement: YOLOAchievement: Quickdraw

Block or report liteshperumalla

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
liteshperumalla/README.md

AWS Docker Kubernetes Terraform

Databases

Postgres MongoDB Redis Oracle

Tools & Others

Git GitHub Jupyter Notebook Visual Studio Code

Recent Projects

Infra Mind – AI Powered Infrastructure Platform

• Engineered an AI driven infrastructure advisory platform with Intelligent dashboards and real–time analytics, executive summaries, automated infrastructure assessments, multi cloud strategy, vendor lock-in analysis, migration scenario modeling, approval workflow system, budget forecasting, rollback automation, security compliance monitoring and reporting capabilities. • Implemented sophisticated data aggregation algorithms, real-time monitoring systems, automated alert mechanisms, comprehensive audit trails, enterprise grade security measures. • Architected a comprehensive multi-agent system with 10+ agents and built a workflow for infrastructure provisioning, compliance monitoring.

Smart Tutor AI

• Led the design and development of an AI-powered tutoring system that delivers course-specific responses using validated academic materials, enhancing learning Faithfulness and relevance by semantic chunking, Hybrid Retrievers. • Developed an efficient retrieval pipeline using metadata tagging and vector search to improve content discoverability and user interaction. • Implemented a multi-agent architecture using LangGraph to orchestrate modular agents responsible for student ID capture, query classification, course-specific routing, and answer clarification—ensuring context-aware, personalized educational support. • Built a Streamlit-based multi-page user interface with authentication (email, Google login), per-user chat history, feedback collection and a customizable course homepage with announcements and professor links. • Integrated LangSmith and LangFuse for real-time system evaluation, enabling continuous fine-tuning and reliability improvements.

MCP Orchestration on GCP

  • LLM/agent initiative using Model Control Protocol to orchestrate tools and data
  • Cloud-native architecture on Google Cloud Platform
  • Tech Stack: GCP, Vertex AI, Cloud Run, Docker, Python

Crypto Portfolio Forecasting

  • ML pipeline with feature store and backtesting capabilities
  • Time series forecasting with advanced feature engineering
  • Tech Stack: Python, pandas, scikit-learn, Feature Store, Streamlit

GitHub Stats:

Litesh's GitHub stats Top Languages

Connect with me:

LinkedIn Email Resume

⭐️ From liteshperumalla

Pinned Loading

  1. Smart-Tutor-AI-AI-Driven-Personalized-Teaching-Support Smart-Tutor-AI-AI-Driven-Personalized-Teaching-Support Public

    With the rise of digital learning platforms, students rely on search engines and LLMs for quick information. However, these tools often return irrelevant results. Smart Tutor AI (STA) bridges this …

    Jupyter Notebook 1 6

  2. Skin-Cancer-prediction-using-machine-learning Skin-Cancer-prediction-using-machine-learning Public

    This project predicts skin cancer using Convolutional Neural Network.

    Jupyter Notebook

  3. Real-time-carbon-emissions-and-weather-impact-Dashboard Real-time-carbon-emissions-and-weather-impact-Dashboard Public

    This is repository is created to share the code for the submission of Data dev Hackathon 2025. This repository contains the code used for this project and the data file used in this hackathon.

    Python 1

  4. Hirefit Hirefit Public

    HIRE FIT is a Big Data and Machine Learning-powered platform that automates resume screening and predicts candidate-job fit using Hadoop, Hive, Amazon S3, AWS SageMaker, and an XGBoost model traine…

    Jupyter Notebook

  5. Powering-AI-Infrastructure-at-Scale Powering-AI-Infrastructure-at-Scale Public

    AI-driven infrastructure advisory platform with intelligent dashboards, real-time analytics, automated assessments, multi-cloud strategy, migration modeling, approval workflows, budget forecasting,…

    Python 1