About
A highly motivated and skilled student with a strong background in Chemical Engineering and a passion for Machine Learning, Generative AI, Transformers, Large Language Models, and Neural Networks.
Work
EarnIn
|Data Science Intern
Summary
Machine Learning Engineer Generative AI, Transformers, Large Language Models, Neural Network
Highlights
Enhanced an MLP classifier with BERT embeddings using Pytorch's embedding layer, improving the F1 score by 10%
Tailored end-to-end pipelines for creating geographic point embeddings using Deep Walk with Word2Vec's skip-gram approach
Engineered custom PyTorch with Petastorm training pipelines for complex neural networks, handling up to 10 million inputs
Volunteer
IIT BHU
|Research Intern
Highlights
Developed a Machine Learning based multiuser localization in UAV-IoT networks using Reconfigurable Intelligent Surfaces
Incorporated Trilateration based techniques to estimate the coordinates of user equipment's followed by joint clustering of the network with K-means Clustering algorithm coupled with elbow method, achieving a localization accuracy of 97 percent
Utilized channel modelling to develop an SINR radio map, trained an RL agent to reach the highest SINR location using various reinforcement learning algorithms including Proximal Policy Optimization, Deep Q Network and Actor Critic Methods
Achieved exceptional SINR optimisation and Interference minimization at RIS, trained in a custom gym environment
Improved the SINR at base station by meticulously adjusting weight vectors and phase of RIS using capon beam forming
Education
Indian Institute of Technology Kharagpur
Bachelor of Technology
Chemical Engineering
Grade: 8.5
Awards
Amazon ML Summer School
Awarded By
Amazon
Earned a spot in Amazon ML Summer School among 85,000 candidates and completed a 4-week ML curriculum
Amazon ML Challenge
Awarded By
Amazon
Attained an overall 8th position finish among 75,000+ teams in the Amazon ML Challenge conducted Pan India
Flipkart Grid 6.0
Awarded By
Flipkart
Qualified for Level 2 and ranked among the top 0.002% (top 328 teams out of 1.6 lakh participants) in Flipkart Grid 6.0.
General Championship Hardware Modelling
Achieved 2nd position in General Championship Hardware Modelling by designing a robust incorrect body posture detector
Publications
Languages
English
Native speaker
Hindi
Native speaker
Skills
Languages
Python, C, C++, SQL, AVRC.
Developer Tools
Spark, Petastorm, Databricks, AWS, VS Code, URDF, Eagle, EasyEda, Tina-TI, Git, Arduino-IDE, Thonny.
Technologies/Frameworks
Linux, GitHub, Tensorflow, Pytorch, Keras, Scikit learn, OpenCv, Hugging Face, BeautifulSoup.
Interests
Social and Cultural
Proficient in several instruments including guitar, keyboard and drums, Performed several performance at school and college level as an instrumentalist and a vocalist, achieving several milestones, Received Third place finish among intra hostels competitions.
Projects
Hangman Challenge
Summary
Transformer Models, Natural Language Processing
Dialogue Summarization using Fine-Tuned LLM
Summary
NLP, Quantization Techniques, Fine Tuning
AI-Powered Voice Salesbot for Real-Time Customer Interaction
Summary
Retrieval-Augmented Generation
Stock Price Forecasting with GAN and Twitter Sentiment Analysis
Summary
Financial Analytics
Neural Style Transfer in Audio Domain
Summary
Deep Learning, Spectral Analysis, Computer Vision