Kaivon Ahmad
Data Scientist
New York, United States
Summary
I'm a tenacious problem solver who is eager to contribute to fascinating technologies with data driven approaches.
Languages:
English
Favorite Python Packages:
Pandas, Numpy, Scipy, Scikit-Learn, Tensorflow, Keras
Experience
Employment
Bubo, Data Scientist, Switzerland (Remote) Nov 2017 ‐ Current
- mined the API's of Youtube, Spotify, Apple Music, etc. to determine market size of particular startup venture observing user listener/views data
- building unsupervised models to extract sentiment with scraped comments data
- collaborating in team of 4
Airmed, Data Analyst, San Francisco Nov 2017 ‐ Jan 2018
- developed data cleaning script with Python to process accelerometer data from wearable sensors
- collaborated in a team of 3 with git version control
Soleeva, Associate, San Jose, CA Apr 2017 ‐ Nov 2017
- designed a proposal tool that interactively visualizes and summarizes the economic benefits of going solar for homeowners
UCSF Diabetes Center, Staff Research Associate II, San Francisco, CA Sep 2016 ‐ Sep 2017
- analyzed gene expression from DNA microarray
- developed different clustering models
- created interactive heatmaps using plotly
Projects
Parkinson's Disease Digital Biomarker DREAM Challenge Aug 2017 ‐ Sep 2017
- extracted features from raw sensor (accelerometer, gyroscope, and magnetometer) time series data recorded during the performance of pre-specified motor tasks obtained from research-grade wearable sensors, smartphones
- developed a machine learning model predict the diagnosis of Parkinson's Disease with an accuracy of 94%
Observing Similarities in Tumor formation from Cap Analysis Gene Expression Jul 2017 ‐ Aug 2017
- explored overlapping gene expression profiles between various types of tumors, both benign and malignant
- extensive wrangling of data acquired from Cap Analysis Gene Expression (CAGE) from human samples by the FANTOM5 consortium
- utilized dimensionality reduction through Principal Component Analysis (PCA) and applied different clustering models including K-Means and Agglomerative
Skills
Amazon Web Services (AWS), Data Science, Git, Machine Learning, Natural Language Processing, NumPy, Pandas, PostgreSQL, SciPy, Spark, TensorFlow, Visualization, Web Scraping