Teja Ijjada

About Teja Ijjada

I am a dedicated Machine Learning professional specializing in computer vision. During my tenure as a Machine Learning intern at Ybi Foundation,I honed my skills in predictive modeling and data preprocessing techniques. Notably, I spearheaded a project aimed at predicting and filtering preferences on movies according to the user\'s choices, achieving notable success using collaborative filtering and content-based filtering, boasting a commendable accuracy of 88%. Developed an AI-powered healthcare assistant that integrates Machine Learning (ML), Deep Learning (DL), and Natural Language Processing (NLP) to assist in disease diagnosis, health risk prediction, and medical consultations->Key Features-Medical Image Diagnosis – Used CNNs (ResNet, VGG16, EfficientNet) for detecting diseases from X-rays, MRIs, and CT scans-AI Chatbot for Medical Consultation – Integrated OpenAI GPT & BERT-based NLP models to provide symptom-based medical advice-Health Risk Prediction – Developed ML models (Random Forest, XGBoost, Logistic Regression) to predict diabetes, heart disease, and stroke risks based on patient data-Voice-enabled Virtual Assistant – Implemented speech recognition for hands-free medical queries and AI-driven responses-Web-Based Interactive Dashboard – Built a Flask/Streamlit app for users to upload images, check diagnoses, and get AI-generated healthcare recommendations->Tech Stack: AI/ML/DL: TensorFlow, PyTorch, OpenAI GPT, Scikit-learn, OpenCV Web Development: Flask, Streamlit, React.js Cloud & Deployment: AWS (EC2, Lambda, S3), Docker, Firebase Impact: Provides AI-driven early disease detection for patients and doctors. Improves medical accessibility with 24/7 AI chatbot consultations. Scales as a telehealth solution for hospitals and clinics With a strong foundation in Machine Learning, Deep Learning and LLM(Large Language Models) coupled with proven achievements in developing innovative solutions, I am poised to make meaningful contributions to cutting-edge projects in Artificial Intelligence.

Experience

1. Machine Learning Engineer

   Insertsol

   Dec 2024 — Present · Visakhapatnam, IN

   Designed and developed an AI-powered adaptive traffic signal control system using real-time video analytics todynamically optimize traffic flow across multi-lane urban intersections.• Implemented YOLOv8 for vehicle detection and integrated vehicle tracking using DeepSORT to estimate real-time trafficdensity, queue length, and vehicle waiting time across multiple lanes.• Built a traffic prediction module using LSTM networks to forecast short-term traffic congestion patterns based onhistorical vehicle flow data.• Developed a reinforcement learning-based traffic controller using Deep Q-Network (DQN) that dynamically adjusts signaltimings to minimize congestion and reduce vehicle waiting time.• Integrated the system with Simulation of Urban Mobility (SUMO) to simulate multi-intersection traffic scenarios andevaluate AI-based adaptive signal strategies.• Reduced average vehicle waiting time by 35% and queue length by 28% in SUMO simulations compared to fixed-timetraffic signals.• Built visualization dashboards using OpenCV and Streamlit to monitor traffic density, predicted congestion, and signaldecisions in real time.

Education

• Dr.Lankapalli Bullayya College

  Bachelor of Technology - BTech

  2020 — 2024