You've made up your mind to become a data scientist. You've taken every data science MooC, you've eaten a lifetime of pizza at machine learning meetups, you even attended a data science "academy." Why hasn't it worked?
Data Science is not knowledge to be acquired but rather a skill that can be learned and improved through practice. The number one qualification employers look for when hiring a data science candidate is previous experience.
Startup.ML is launching a fellowship to give aspiring data scientists the chance to hone their skills by building real machine learning applications for startups and established data science teams. Upon completing the program, fellows will have direct access to a network of hiring partners, a letter of recommendation, and a portfolio of real world projects.
A Fellowship Designed to Maximize Practical Experience
Full time for 4 months at our San Francisco location
Fellows build scalable machine learning models that integrate into real products
We follow an agile development process in groups of 3 (pair programming plus agile team lead). This means weekly iteration planning, daily scrum and every Friday is demo day!
We strive to have each fellow work on two separate projects to give them exposure to a wider range of problems.
Work on a Project for an Established Data Science Team
We ensure that our fellows work on real world problems sourced through partnerships with established data science teams. This aspect of the work is well defined, datasets have already been gathered, and there is an opportunity to learn from leading practitioners.
The challenge fellows will typically confront is how to get results when working in a large team with lots of differing opinions, approaches, skills and priorities.
Another key skill that fellows learn by working with an established data science team, is the ability to communicate ideas and build consensus. We ask fellows to present their work to outside team members on a weekly basis which hones their ability to deliver crisp results.
Work on a Nebulous Startup Problem
Startup.ML works with early stage companies to help them incorporate machine learning into their products. We afford fellows the opportunity to work on these problems along with us. The learning opportunity for this part of the fellowship is different from the previous experience of working with an established data science team.
There may be a lack of clarity about what exactly needs to be done, the data may not exist or if it exists it often doesn't have the necessary signal, and the founders are uncertain about the direction they want to take their product.
Fellows need to learn how to iterate quickly and have a laser-focus on producing a solid minimum viable product.
Pairing Software Engineers with Quants
Data Science is an interdisciplinary field which combines aspects of computer science, mathematics and statistics. Rarely do we see someone that has a deep background in all of these areas. We encounter software engineers that are not familiar with probabilistic approaches to problem solving and quants (mathematicians, statisticians, physicists, computational biologists, etc.) that can't write a recursive function.
Instead of holding out for the unicorn, we pair fellows from each of these backgrounds to work on a problem together. The pair-programming methodology has already proven to be extremely effective when building software and we believe it’s also the right approach for data science.
While it is true that the problem should dictate the choice of the tool, we have developed some preferences.
Julia is a new but very promising language for scientific computing. We have successfully used Julia for digital signal processing, state-space tracking and anomaly detection.
Fellows are free to pick their choice of tools but using a tool that mentors are familiar with, makes it easier to get help.
After the Program
At the completion of the program, we help fellows find the right career opportunity. Fellows will leave the program with intimate knowledge of the startups they are working with, while also having numerous opportunities to engage with guest speakers from established data science teams.
How to Apply
Our first cohort starts on March 9, 2015. We are accepting applications for future cohorts.