You will architect, code and deploy ML models (from scratch) to predict credit risk.
You will design, run, and analyze A/B and multivariate tests to test hypotheses aimed at optimizing user experience and portfolio risk.
You will perform data exploration and build statistical models on user behavior to discover opportunities for decreasing user defaults. And you must truly be excited about this part.
You’ll use behavioral and social data to gain insights into how humans make financial choices
You will spend a lot of time in building out predictive features from super sparse data sources.
You’ll continually acquire new data sources to develop a rich dataset that characterizes risk.
You will code, drink, breathe and live python, sklearn and pandas. It’s good to have experience in these but not a necessity - as long as you’re super comfortable in a language of your choice.
About you:
You’ve strong computer science fundamentals
You’ve strong understanding of ML algorithms
Ideally, you have 2+ years of experience in using ML in industry environment
You know how to run tests and understand their results from a statistical perspective
You love freedom and hate being micromanaged. You own products end to end
You have a strong desire to learn and use the latest machine learning algorithms
It will be great if you have one of the following to share - a kaggle or a github profile
Degree in statistics/quant/engineering from Tier-1 institutes.