Jun 2024 – Mar 2025 · 9 mos · Bengaluru, Karnataka, India · Hybrid

Jul 2022 – Jun 2024 · 1 yr 11 mos · Gurugram, Haryana, India

• 1. Collaborated in development of a refurbishment cost estimation system using XG Boost for classification and an Extra Tree Regressor for cost prediction. By analyzing comments from CJ, the system accurately determined projected expenses during automobile inspections, enabling informed pricing decisions.
• 2. Trained a Faster RCNN with a ResNet-50 model featuring FPN (Feature Pyramid Network) for accurate detection of different panels of cars. The output of this detection was then fed into another ResNet model for further classification, determining whether each detected panel was damaged or intact.
• 3. Developed a robust classification model using Extra Tree Classifier that leverages fingerprint-level features, including age, odometer reading, CJ behavior, previous inspection misses, and over-reporting. The model accurately predicts whether a CJ comment corresponds to an inspection miss, over-reporting, or a correct report.
• 4. Conducted an in-depth analysis of engine sounds by applying the Fast Fourier Transform (FFT) to extract frequency domain information. Leveraging the obtained features, developed a classification model to assess the health of the engine, accurately determining whether it is in good condition or requires attention.
• 5. Undertook a critical project to refactor legacy code, optimizing its performance and enhancing time complexity significantly. The revamped code was then redeployed, leading to a more efficient and scalable workflow. The machine learning part of the application was containerized using Kubernetes for enhanced manageability, while the entire codebase was maintained and deployed as a Google Cloud Platform (GCP) cloud function.

Dec 2021 – Jun 2022 · 6 mos · Gurugram, Haryana, India

• Worked on detection and classification on damages such as scratch , rust, dent etc on car using computer vision.
• Helped business procure good condition cars without any manual labour on exterior parts or cars.
• Rejected cars decreased by 50%.

Aug 2021 – Nov 2021 · 3 mos · Chennai, Tamil Nadu, India

• Working as an intern at one of the most prestigious Institute of India - Indian Institue of Technology- Madras
• Supervisior - Krishnan Balasubramanian
• Topic - Locate detonators and bombs using CT Scanned Images.
• Domain - Computer Vision

Jun 2021 – Aug 2021 · 2 mos · Allahabad, Uttar Pradesh, India

• Worked as an intern at CCC lab, IIIT Allahabad
• Supervisor - U.S Tiwari, Rohit Mishra
• Topic - Dialogue State Tracking
• Domain - Natural language processing, Dialogue System

Feb 2021 – Oct 2021 · 8 mos · Bangalore Urban, Karnataka, India

• 1. Successfully completed a range of projects, encompassing Language Identification, Invoice Classification, Address Validation, OCR Testing, and OCR for Invoices.
• 2. Acquired advanced proficiency in Natural Language Processing (NLP) topics, such as Transformers, BERT, and Embeddings, through hands-on experience and project work.
• 3. Demonstrated the ability to apply acquired skills to real-world business scenarios and projects, showcasing practical competence in NLP and related domains.

Jul 2020 – Sep 2020 · 2 mos

May 2020 – Jul 2020 · 2 mos