My favorite language is Python, but I always enjoy learning new things and trying out new languages/frameworks. I have 4+ years of Python experience from all three of my internships as well as my coursework, research and personal projects. I have experience with all kinds of Python libraries from using sklearn for machine learning to Django/Flask for the web.

I have full stack web experience from my internships at Khan Academy and Bevspot as well as personal projects (like this one) and from coursework. On the back end, I've mostly used Flask/Django as well as Node.js and some Meteor.js. On the front end, I've used mostly standard HTML/CSS/Javscript as well as a bit of React.

Java used to be one of my primary languages, but I don't really use it much anymore. I've messed around with some C and C++ in internships and some on my own but not regularly. I can also cook up some mean bash scripts when the time comes.

For version control, I'm familiar with git using various common workflows. I prefer Sublime or Vim for writing code. I could continue listing tools I've used over the past 5-6 years, but instead I'll throw in a plug for my L^aT_eX experience by letting you view my resume.

For my Master's of Engineering, I'm focusing on machine learning. My thesis focuses on applying machine learning and Natural Langauge Processing (NLP) to medicine working under Professor Regina Barzilay at MIT CSAIL and Dr. Charlotta Lindvall at the Dana Farber Cancer Institute in Boston. While at MIT, I have taken multiple ML courses including an advanced NLP course as well as multiple probability/statistics courses.

For my M.Eng. thesis, I have used NLP and ML algorithms to predict the effectiveness of a treatment for heart failure using electronic health records. Up to 90% of electronic health record data is in free text, doctors' notes, format requiring the use of NLP for effective utilization. My work so far has increased the precision of predicting non-response for a heart failure treatment from 57.9 to 89.5% compared to the currently used method. Now, I am focusing on extending the techniques developed in order to apply them to a wide range of medical problems.

Beyond ML, I have a strong algorithms background from coursework as well as some algorithms research experience from my internship at Akamai. At MIT, I took the intro algorithms course as well as one more advanced followup course. At Akamai, I designed and implemented an algorithm forming the theoretical basis of a new type of content distribution network that would be used in next generation network architectures.

Both from coursework and from my Akamai internship, I have experience with computer systems engineering and distributed systems. This includes familiarity with ACID properties, security, and real world system design such as Internet protocols, databases, and common encryption schemes.

As a data science/software engineering intern at Khan Academy, one of my projects involved building infrastructure to collect site health metrics and to alert upon statistically significant failures. Using both properly selected probabilistic models and feature engineering, I developed a monitoring system without an unnecessarily high false positive rate. In addition to that project, I worked alongside the infrastructure and data science teams on every day tasks across their full web stack doing everything from bug squashing to small feature implementation to improving test coverage.

A large part of my M.Eng. thesis so far has been implementing NLP and ML algorithms on real electronic health record data. For this, I primarily use Python/sklearn. Going forward, at least half of the work will be implementation focused.

While a lot of the work I did at Akamai was theory focused, I also implemented a poly-time approximation scheme for one of the algorithms I developed and applied it to production data.

During my short, one month winter "externship" at Bevspot I worked across their full web stack. My role was primarily back end/infrastructure focused, replacing part of their data model and developing a new concurrent data structure for their web application. I also worked on a variety of other small front and back end tasks throughout the externship

My first "practical" experiences came from research. I did a UROP at MIT in computational biophysics, and I did computational biology research for two years in high school at the Mayo Clinic and the University of North Florida. At MIT, I developed Python tools to predict protein-DNA interactions and designed a new protein-DNA docking algorithm. In my high school research project, I primarily sought to increase the efficiency of an algorithm used for drug discovery, and used it to identify potential drug candidates for treating cancer.

Outside of all my work experience, I sometimes work on some of my own projects. You can see more details about these below.

My primary interest is machine learning and artificial intelligence. I am interested largely in the applications of relatively simple concepts in ML to solving real world problems such as in medicine.

As a software engineer, my goal is to write maintainable, efficient code that gets the job done. I enjoy tackling hard problems and getting things done.