About

Aspiring Machine Learning Engineer with hands-on experience in Machine Learning, Deep Learning, Data Science, NLP, Computer Vision, and Audio AI. Skilled in Python, PyTorch, TensorFlow, Scikit-learn, feature engineering, model optimization, and predictive analytics. Experienced in developing end-to-end AI solutions, audio deepfake detection systems, and data-driven applications using real-world datasets.

Experience3

1. VC

   Data Analyst

   Jan 2026 — Present

   Vizva Consultancy Service · Ahmedabad Gujarat, India

   Built and delivered interactive dashboards using Microsoft Power BI to track KPIs, improving visibility of business performance for stakeholders. Performed data preprocessing and exploratory data analysis (EDA) using Python (Pandas, NumPy) to uncover trends and prepare clean datasets for reporting and modeling. Developed machine learning models for client use-cases such as prediction and trend analysis, with proper evaluation using standard metrics. Translated business requirements into data-driven solutions by collaborating with clients and presenting insights through clear visualizations and reports.

   • Improved visibility of business performance for stakeholders through interactive dashboards.
   • Uncovered trends and prepared clean datasets for reporting and modeling using EDA.
   • Developed machine learning models for prediction and trend analysis.
2. IS

   Data Science Intern

   May 2025 — Jul 2025

   Indian Statistical Institute, Kolkata · Internship · Kolkata, India

   Completed a structured 32-hour training program on Power BI, Streamlit, Machine Learning, Deep Learning, Generative AI, and Business Intelligence with LLMs. Built predictive models for stock market movement using sentiment analysis and time-series features. Carried out data cleaning, feature engineering, and exploratory data analysis on financial datasets. Applied machine learning algorithms and evaluated performance using accuracy, precision, and recall.

   • Built predictive models for stock market movement.
   • Completed a structured training program on various AI technologies.
3. P

   Machine Learning Intern

   Feb 2023 — Apr 2023

   Pregrad · Internship · Remote

   Completed industry-led training on Machine Learning fundamentals and real-world applications. Applied supervised and unsupervised learning techniques for data preprocessing, model development, and evaluation. Worked on project-based ML solutions using Python, Scikit-learn, Pandas, Matplotlib, and Jupyter Notebooks.

Projects4

• AI Agent for Task Automation

  Building an AI-based personal assistant capable of automating routine tasks like email replies, meeting scheduling, and cab bookings. Implemented basic intent detection and entity extraction using NLP techniques. Integrated task handlers for replying to emails and scheduling based on extracted time, date, and subject. Currently working on enhancing precise location extraction using contextual NLP and NER.

  • Python
  • NLP
  • NER

• ArtifactNet: Audio Deepfake Detection

  Designed an interpretable audio deepfake detection framework for ASVspoof 2019 LA and ASVspoof 2021 datasets, focusing on synthesis artifacts rather than speaker identity cues. Engineered a feature extraction pipeline combining dynamic spectral artifacts, LPC residuals, and frame-level frequency inconsistencies from over 25,000 audio samples. Developed a lightweight Bi-GRU encoder with attention pooling for temporal modeling of spoofing artifacts. Evaluated performance using AUC, EER, and t-DCF metrics with cross-dataset validation.

  • Python
  • PyTorch
  • Signal Processing
  • Deep Learning

• Audio Anonymization – MC-SEA

  Developed a speaker anonymization framework using margin-controlled speaker embedding replacement without training additional neural networks. Extracted ECAPA-TDNN speaker embeddings and performed cosine-distance based matching from an external speaker pool for identity masking. Implemented an efficient anonymization pipeline with significantly lower computational overhead than adversarial and voice-conversion based approaches.

  • Python
  • SpeechBrain
  • ECAPA-TDNN

• Helmet Detection using ResNet-50

  Built a helmet detection system using transfer learning with ResNet-50 on a dataset of 764 annotated road-safety images. Applied image preprocessing, augmentation, and fine-tuning techniques to improve model generalization and reduce overfitting. Implemented an end-to-end binary classification pipeline for automated identification of helmet compliance in traffic scenarios.

  • Python
  • TensorFlow
  • Computer Vision

Education1

• AU

  B.Tech in Computer Science & Engineering (AI & ML)

  2022 — 2026

  Adamas University

Certifications6

• Microsoft Certified: Azure AI Fundamentals

• Microsoft Fundamentals of Computer Vision

• Build Real World AI Applications with Gemini and Imagen, Google Cloud Skills

• Introduction to Docker, Google Cloud Skills

• Prompt Design in Vertex AI, Google Cloud Skills

• NPTEL Problem Solving through Programming in C