Data Engineer
2019 - 2020Hart Inc.- Designed and developed an ML-based health data search engine via automated schema prediction.
- Modified existing databases to meet unique needs and goals determined during initial evaluation and planning process.
- Wrote scripts and processes for data integration and bug fixes in Python, Scala, and Java.
- Planned, engineered, configured, and deployed ML tooling and big data solutions while supporting software in a Hadoop-Spark ecosystem.
Technologies: Apache Airflow, Scala, Python, Spark, Hadoop, PySparkTechOps Engineer
2018 - 2019Telescope Inc.- Built business logic for voting applications and directly support the world's largest live shows such as The Voice, American Idol, and Dancing with the Stars.
- Advised and provided versatile big data solutions using AWS (Dynamo, S3, EC2, etc.) to meet needs of clients.
- Wrote unit tests in Python to automate product validation.
- Researched and developed the integration of smart home devices to the current platform, provided prototypes as a proof of concept to executives/clients, and improved profit margins by launching new add-on projects for existing clients.
Technologies: Amazon Web Services (AWS), Apache Kafka, Flume, REST APIs, RESTful Development, XML, SQL, JavaScript, Spark, PythonDevelopment Engineer/Engineering Consultant
2016 - 2018UCLA, Various Startups (Vortex Biosciences Inc., Ferrologix Inc., etc)- Wrote code to automate statistical analysis on vision data (live/recorded microscopic images/videos) using data science and computer vision tools in Python, MATLAB, and R.
- Trained employees on usage of aforementioned codes remotely.
- Addressed R&D issues from data mining perspective in developing microfluidics platforms for medical applications.
- Authored publications in peer-reviewed scientific journals.
- Consulted start-up companies and assist director with project management.
Technologies: Amazon Web Services (AWS), R, MATLAB, Linux, Tableau, SQL, PythonRobotics Researcher
2014 - 2016UCLA- Developed algorithms to be tested on custom humanoid platforms using Simulink, Python, C++, Lua, ROS, LabView, and COMSOL.
- Maintained robot platforms using CAD, 3D rapid prototyping and CNC mill.
- Competed in DARPA Robotics Challenge Final as Team THOR. (USA, Jun 2015).
- Competed in RoboCup as Team THORwIn (China, Jul 2015) – 1st place winner in the adult-sized humanoid open platform.
- Work as robotics education outreach activity coordinator.
Technologies: Amazon Web Services (AWS), Java, Robot Operating System (ROS), C++, R, MATLAB, Python
Rachel Park
Software Developer in Los Angeles, CA, United States
Member since April 21, 2020
Rachel is a big data professional experienced in various domains including robotics, biotech R&D, entertainment/media, and healthcare. With 8 years of experience in technologies for data mining and machine learning, she's a proactive leader with strengths in communication and collaboration. She is proficient in leveraging the Hadoop-Spark ecosystem, using ML knowledge to promote automated data solutions, and managing concurrent objectives to promote efficiency and influence positive outcomes.
Rachel is now available for hire
Portfolio
- Hart Inc.Apache Airflow, Scala, Python, Spark, Hadoop, PySpark
- Telescope Inc.Amazon Web Services (AWS), Apache Kafka, Flume, REST APIs...
- UCLA, Various Startups (Vortex Biosciences Inc., Ferrologix Inc., etc)Amazon Web Services (AWS), R, MATLAB, Linux, Tableau, SQL, Python
Availability
Full-time
Preferred Environment
Sublime Text, Anaconda, Atom, PyCharm, Visual Studio Code
The most amazing...
...thing I've developed is a machine/deep learning based automatic query generator tool. It utilizes Spark MLLib and Keras.
Employment
Experience
- Data Type Predictor
Utilizing Spark MLLib, it predicts data types of fields in a given database. Useful when the inferred data type is not accurate or insufficient. Also, it serves as a key to finding primary key and foreign key relationship.
- Semantic Type Predictorhttps://sherlock.media.mit.edu
This is a PySpark implementation of the deep learning tool Sherlock.
This implementation predicts the semantic type of fields in a given database. It is useful for cleaning data and matching schema. - Single Cell Image Identifier
Given a blood sample provided by terminal cancer patients, the goal is to identify circulating tumor cells from various other cells (mostly white blood cells) and debris in the blood by analyzing the morphology of these cells and applying machine learning techniques.
Skills
Languages
Python 3, SQL, Python, XML, Scala, Java, JavaScript, R, C++Frameworks
Spark, Hadoop, DjangoLibraries/APIs
Spark ML, PySpark, REST APIsTools
Apache Airflow, Spark SQL, MATLAB, PyCharm, Atom, Sublime Text, Flume, TableauParadigms
ETL, RESTful DevelopmentPlatforms
Amazon Web Services (AWS), Docker, Kubernetes, Visual Studio Code, Linux, Databricks, Anaconda, Apache KafkaStorage
MySQL, PostgreSQLOther
Robot Operating System (ROS), Machine Learning, Deep Learning
Education
- Master of Science Degree in Mechanical Engineering2014 - 2016University of California, Los Angeles - Los Angeles, CA
- Bachelor of Science Degree in Biomedical Engineering2009 - 2013Johns Hopkins University - Baltimore, MD
Certifications
- Big Data Hadoop CertificationJANUARY 2020 - PRESENTEdureka
