Developed an LLM for document question-answering on SageMaker Jumpstart using the FLAN-T5 model, which was parameter fine-tuned and instruction fine-tuned using domain-specific PDF files.
Performed conversion of code of a top Kaggle submission for horse race winner prediction from R to Python, such that it’s designed to work with the client’s own dataset.
Executed an SQL query on AWS Glue, and then performed Overall Equipment Effectiveness (OEE) calculation for different machines, parts, tools and their combinations. Then developed an ML model for one particular machine-tool combination for predicting total scrap, which gave an R2 score of 0.84
Worked on a defect detection object detection Rekognition Custom Labels project, involving data curation, data labelling, model training and evaluation, and running inference using AWS Lambda.
Education
Bachelor of Engineering - Electrical And Electronics Engineering
Birla Institute of Technology And Science, Pilani
Hyderabad, TG
01-2022
Skills
Python
AI/ML
AWS
R
SQL
Flask
HTML, CSS
Personal Projects
Food Delivery Time Prediction:
Performed Data Cleaning by dealing with hidden null values, converting data types, and
dealing with some incorrect order pick dates.
Built an XGBoost model for predicting food delivery time, which returned with a mean
squared error of 16%.
Bank Customer Churn Prediction:
Performed EDA and model building for churn prediction.
Used SMOTE technique to deal with imbalanced dataset, and finally return an XGBoost
classification accuracy score of 87%.
Deployed the ML model using Flask.
Certification
AWS Certified Machine Learning Specialty
Languages
Telugu
First Language
English
Proficient (C2)
C2
Hindi
Upper Intermediate (B2)
B2
Japanese
Elementary (A2)
A2
Timeline
Machine Learning Engineer
Minfy Technologies Private Limited
09.2023 - Current
Bachelor of Engineering - Electrical And Electronics Engineering