Data Scientist Resume

TECHNOLOGY:

Python MATLAB AFNI/SPM SPSS LaTeX MAX/Msp Git/GitHub SQL/noSQL Hadoop/Spark AWS Unix/Linux Jupyter TensorFlow NLTK word2vec

PROFESSIONAL EXPERIENCE:

Confidential

Data Scientist

Responsibilities:

12 week intensive accredited bootcamp focused on enhancing data science skills.
used Python, tools for data acquisition (web scraping/APIs), data management (AWS, MongoDB, SQL, Hadoop), statistical analysis, machine learning (supervised and unsupervised), natural language processing, and data visualization. Select projects highlighted below.

Confidential, Baltimore

Faculty

Responsibilities:

College Biology, Introduction to Computer Science, Foundations of Technology, College Statistics, and Survey of Music.
Trained faculty and staff to proctor and administer state/national exams.

Confidential

Postdoctoral Fellow

Responsibilities:

7 peer - reviewed publications using non-linear signal processing of auditory stimuli and functional neuroanatomy.
Applied machine learning and statistical techniques to musical, behavioral, and fMRI data.
PCA, ICA, wavelet analysis, multivariate statistical signal processing and estimation methods, Monte Carlo methods, auto-regressive models, co-dimensionality.
Supervised and unsupervised models (SVM), regression models, circular statistics.
Formulation and implementation of experiments designed to address specific questions.
Complex technical setup (including audio equipment, MRI machines, audio booths) and stimulus creation (ePrime, Labview, MAX/Msp, Logic Pro) for human cognitive, motor and behavioral studies in developmental, diseased and normal populations.
Invited public talks for academic, student and lay audiences.
Mentored Hopkins engineering and medical students in experimental design, methods, and analysis.
Created and managed detailed documentation schemes and standards of reproducibility. (database of subjects and related IRB materials for multiple labs).

Confidential

Graduate Student, System Administrator

Responsibilities:

Upkeep of hardware (computer, audio), software, networking, website and purchasing equipment.
Built a TED talk recommender using Natural Language Processing (NLP) on transcripts of talks.
Recommendations based on topic modeling with Latent Dirichlet Allocation (LDA).
Matching a score to a performance to extract errors, tempo, and beat structure.
Classified sounds from cats and dogs (audio wav files) using time-frequency analysis, dimensionality reduction (PCA) and supervised classification models.
Calculated relationships between health factors contributing to obesity rates (CDC datasets) and locations of Confidential (web scraping) through regression models (linear L1, crossvalidation).