top of page

If you're here because you have watched my video, I'm glad it worked!
Don't know what I'm talking about? Watch it below!


👋Hi, I'm Abhilasha!

📍 Boston, Massachusetts, United States

Creative and collaborative engineer with 5+ years of experience in Data Science and Machine Learning. Proactive team player with a find-the-solution attitude.

Demonstrated ability to work independently in multicultural and fast-paced environments.

Experienced working in and leading cross-functional teams, including software development, deployment, data engineering, and analysis.


🛴 Quickly Adaptable to a new environment (Very Flexible)

🤓 Self Starter and Fast Learner

💬 Excellent Communicator

⌛ Get-It-Done Attitude

✅ Problem Solver

Technical Proficiencies

  • Python

  • SQL

  • R

  • C/C++


  • Tableau

  • Power BI

  • Salesforce

  • Azure

  • Machine Learning

  • Deep Learning

  • Image Processing

  • Signal Processing

  • Generative AI

  • Computer Vision

  • Statistical Modeling

  • Mathematical Modeling

💻3 years as Data Scientist Product Manager

For 3 years, I led and mentored cross-functional teams in prototyping products as per client's requirements.

I managed projects from initiation to completion, ensuring ethics and compliance guidelines were consistently met. I collaborated with multiple stakeholders to deliver actionable insights and reports for business decision-making processes. 

🛠️ Created Internal Tools 

  • Developed an automated reimbursement system using Optical Character Recognition, Named Entity Recognition, and Prompt Engineering reducing manual labor by 73% and time consumption by 65%

  • Fine-tuned multiple open-source LLMs and developed ETL pipelines with Azure

  • Automated data annotation tasks, increasing work efficiency by 30% and reducing errors

🔺 Increased Sales by 20% in a month

  • Applied advanced Machine Learning techniques, including Large Language Models (LLMs) and Generative AI to develop predictive models for customer behavior analysis and product recommendation systems, resulting in a 20% increase in sales

  • Designed and developed NLP algorithms to extract insights from unstructured text data, leading to a better understanding of customer behavior, leading to improved decision-making process.

🖥️ Designed Novel Biomedical Solution during COVID-19 

  • Initiated the development of multi-patient ventilator valves during the pandemic to deal with the ventilator shortage.

  • Later, the initiative was overtaken by a large company with better resources.

📈 Reduced Data Processing Time by 40% and Deployment Time by 30%​

  • Deployed Flask-based RestAPI for real-time ML workflow, reducing model deployment time.

  • Implemented Random-forest based predictive algorithms, improving Heart Attack prediction accuracy by 20%.

  • Developed and maintained automated data collection and cleaning pipelines, reducing data processing time by 40% and improving data quality 

Notable Projects
Skin Lesion Melanoma Classification
  • Developed a Convolutional Neural Network (CNN), leveraging architectures like ResNet and VGG, to accurately classify skin lesions and distinguish melanoma from benign lesions.

  • Improved classification accuracy by implementing precision and recall metrics, resulting in reduced false positives and false negatives, thus enhancing early detection and potentially saving lives.

Biosignal Denoising on Apnea Dataset
  • Employed Fast Fourier Transform (FFT) and Adaptive Filtering techniques, such as Kalman filter on Hankel Matrix, to enhance the quality of biosignals collected from apnea patients by reducing noise and artifacts.

  • Improved Signal-to-Noise Ratio (SNR) and decreased Mean Squared Error (MSE), leading to more reliable diagnosis and monitoring of apnea conditions, thereby improving treatment outcomes.

Pulse Diagnosis Mathematical Model
  • Developed a mathematical model based on machine learning algorithms such as Support Vector Machines (SVMs), Decision Trees, and Artificial Neural Networks (ANNs) to assist practitioners in traditional pulse diagnosis.
  • Enhanced diagnostic accuracy by utilizing accuracy, sensitivity, and specificity metrics, providing practitioners with a systematic and objective tool for treatment planning and decision-making.

​Stock Market Analysis
  • Utilized Autoregressive Integrated Moving Average (ARIMA) and Long Short-Term Memory (LSTM) networks for time series analysis to predict stock market trends.

  • Reduced Mean Absolute Error (MAE) and Mean Squared Error (MSE), providing investors with more reliable insights for informed investment decisions.

Vehicle Tracking and Speed Detection on Surveillance Camera
  • Implemented YOLO object detection and Optical Flow for motion estimation to accurately track vehicles and enforce speed limits on surveillance camera footage.

  • Improved accuracy of vehicle detection and speed estimation, enhancing traffic management and safety measures.

Fraud Detection on Digital Transactions
  • Deployed Isolation Forest, XGBoost, and Random Forest machine learning models to detect fraudulent transactions.

  • Increased precision and recall rates, leading to more efficient fraud detection and reduced financial losses for businesses.

Customer Churn Prediction
  • Employed Logistic Regression, Random Forest, and Gradient Boosting Machines (GBM) models to predict customer churn.

  • Improved accuracy, precision, recall, and F1-score, enabling businesses to proactively identify and retain customers, thus reducing churn rates and increasing customer lifetime value.

bottom of page