Sai Teja Jesetti

Experience

• 

  Internshala Trainings

  Jun 2020 - Aug 2020

  Data science trainee

  • Acquired expertise in Data Science through completion of key modules such as Introduction to Data Science, Python for Data Science, Understanding Statistics for Data Science, and Predictive Modeling and Basics of Machine Learning.• Recognized as a top performer during the training program.

• 

  The Microchip Academy

  Sept 2020 - Nov 2020

  Electronics intern

  • Gained hands-on experience with Arduino UNO. Engineered an "Object detection bot" using Arduino UNO, increasing object detection accuracy by 25%

• 

  Internshala Trainings

  Nov 2020 - Feb 2021

  Machine Learning Intern

  • Developed and implemented machine learning algorithms for the Credit Card Defaulters project using Python libraries (Pandas, NumPy, SciPy, Scikit-Learn). Conducted extensive data preprocessing, feature engineering, and optimized model training, achieving an 85% accuracy rate.• Improved model accuracy by 15% through hyperparameter tuning.

• 

  Freelance

  Jan 2021 - Aug 2023

  AI/ML Freelancer

  • Developed and deployed a visual transformer model on AWS, achieving a 22% increase in classification accuracy (from 70% to 92%) for client projects. • Engineered a custom CNN for identity document recognition with a focus on Aadhaar cards and passports, ensuring high accuracy and seamless client integration through a RESTful API on AWS EC2. • Built and deployed predictive models with a 95% precision rate, enabling clients to make data-driven decisions and optimize resource allocation, leading to significant cost savings. • Leveraged cloud platforms (AWS, GCP) for scalable and reliable AI model deployment, ensuring high availability and performance. • Provided technical consultations and support to clients, guiding them on AI/ML integration and best practices. Show less

• 

  BigGraph AI Technologies, Inc

  May 2023 - Sept 2023

  Data Scientist

  • Enhanced AI application performance by 20% through optimization of ChromaDB and Pinecone vector databases, resulting in faster data retrieval and improved user experience. • Fine-tuned state-of-the-art LLM models (Llama2, Mistral, OpenAI) to achieve project-specific performance goals, including improved accuracy, reduced latency, and enhanced contextual understanding. • Developed and implemented NLP pipelines for text classification, entity extraction, and sentiment analysis, contributing to the development of advanced AI-powered solutions. • Reduced search response time by 90% using Whoosh and BERT, significantly improving application efficiency and user satisfaction. • Collaborated with cross-functional teams to understand business requirements and translate them into effective AI solutions. Show less

• 

  Genzeon

  May 2024 - now

  NLP Engineer

  • Working as a NLP Engineer for Genzeon, specializing in machine learning and natural language processing (NLP) projects.• Currently involved in a significant project for a leading healthcare client, focusing on extracting and analyzing data from medical documents.• Developing advanced OCR and NLP models to enhance data extraction accuracy for complex medical documents.• Utilizing frameworks such as TensorFlow, PyTorch, and Keras to build and deploy machine learning models.• Collaborating with cross-functional teams to ensure the successful implementation and deployment of solutions. Show less