Jan 2023 – Jun 2023 · 5 mos · Bengaluru, Karnataka, India · Remote

• Developed a browse node classification system for the catalog team, by finetuning DistilBERT model achieving an F1 score of 92% which was able to classify the product out of 325 classes.
• Implemented ML and DL algorithms (Logistic Regression, SVM, KNN, Decision Tree, BiLSTM, CNN, GRU) to establish a robust baseline, and conducted extensive data preprocessing and feature engineering on real-time data.
• Experimented with different transformer models like BERT, RoBERTa etc. to choose the model best suitable for the problem statement.
• Utilised Back translation method using MarianMT to balance the data
• Reduced headcount requirement for product classification by 5x, optimizing operational efficiency.
• Created a functional prototype with Streamlit for stakeholder demonstration and interactive testing.
• Authored comprehensive documentation, including user guides, technical specifications, and model deployment instructions.

Oct 2022 – Dec 2022 · 2 mos · Vellore, Tamil Nadu, India

• Led the research and development of a sugarcane crop health monitoring robot using computer vision and transfer learning.
• Collected, labeled, and preprocessed over 5,000 real-time data points, showcasing expertise in computer vision and dataset management.
• Contributed to the development of a sugarcane disease classification model.

Jul 2022 – Jul 2022 · 0 mo · Mannar, Kerala

Jul 2021 – Aug 2021 · 1 mo · Tiruvalla, Kerala, India

• I worked in the maintenance department of the company
• https://drive.google.com/file/d/1YRFW2Obkcrbtk_668EMXkG8a_bZ2V2dR/view?usp=drivesdk