cv_oct_2024 (1)
cv_oct_2024 (1)
Bengaluru, Karnataka
[email protected] | github.com/shreyahegde18 | linkedin.com/in/shreya–hegde–/
EDUCATION
Bachelor of Engineering in Artificial Intelligence and Machine Learning
2020-2024
Acharya Institute of Technology | Bengaluru, India
Intermediate/+2 , Jawahar Navodaya Vidyalaya (JNV) | , North Canara (CBSE) 2013-2020
Position of Responsibility: Sports Leader, Coding Club Coordinator
EXPERIENCE
Target, Apprentice Data Science | Bengaluru, India Aug 2024 - Current
• Worked under Research and Development team of Target in India (TII)
• Collaborated with the ongoing projects in the US team
• Contributed to taking ideas to POC level
American University of Ras Al Khaimah, AIML Intern | Remote (UAE) Aug 2023 - Sep 2023
• Collaborated with the Ras Al Khaimah Research and Innovation Centre (RAKRIC).
• Contributed to the development of a Chatbot for mental health support, designed to deliver empathetic and imme-
diate responses to users.
• Implemented a synthetic data generation strategy to enhance the model’s ability to handle a wide range of mental
health-related conversations.
SKILLS
Languages Python, C, Git
Development MySQL, HTML, CSS, Streamlit, Flask
ML Frameworks & Vector Dbs Pytorch, Tensorflow, Scikit-learn, Langchain, Pinecone, Chroma
Soft Skills Commitment
PROJECTS
Computer Auditory Imagination Feb 2024 - Apr 2024
• Developed a multi-model approach to generate visually compelling videos for non-linguistic audio contents.
• Finetuned AST on ESC50 dataset for Audio classification with accuracy of 92.75%. fine-tuned LLM with Prompt
engineering techniques and DiffusionPipeline for sequential video generation
• Tech stack - Python, LLMs, torch, Huggingface Models.
Real-time Sign Language Recognition in Low-light Conditions Jun 2023 - Jan 2024
• Pioneered the creation of a unique dataset of 2000+ low-light hand images for model training, addressing the gap in
available datasets.
• Utilized a tech stack comprising Python, OpenCV, Mediapipe framework and Google Teachable Machine.
• Hyper-parameter fine-tuning to achieve a testing accuracy of 99.00%. The model is computationally efficient,
processing 3 fps in extremely dark environments.
News Research Tool Sep 2023 - Nov 2023
• Built a News Research Tool, a website that extracts news from URLs(can be extended to 50+ URLs) and generates
answers to relevant questions.
• Leveraged Python, Langchain, Faiss-CPU Vectordb, Streamlit, and certain LLMs for the implementation.
• The tool offers a user-friendly interface.
RESEARCH WORK
Region-Specific Adaptive Weighted Multiscale Retinex (RS-AWMSR): A Real-Time
Approach to Low-Light Image Enhancement for Sign Language Recognition
(3rd IEEE International Conference on Computer Vision and Machine Intelligence, IIIT Allahabad, Uttar Pradesh)
Computer Auditory Imagination: A Multi-Model Approach for Audio to Video Synthesis
(International Conference On Generative AI, Cryptography and Predictive Analytics 2024, Vivekananda Institute
of Professional Studies, New Delhi)
ACHIEVEMENTS
Chess Nationals: Participated in National level Chess tournaments held across various parts of India 4 times.