Senior Data Scientist
San Francisco, CA
Full Time - Exempt
Affirm is reinventing credit to make it more honest and friendly, giving consumers the flexibility to buy now and pay later without any hidden fees or compounding interest.
Affirm’s data science team solves problems critical to Affirm’s business model - detecting fraud and assessing creditworthiness in real time. Affirm’s innovative products ensure that we cannot use off-the-shelf modeling algorithms and must create novel data science solutions to drive both existing and new products. At some point during almost every project you work on, you will think: “I thought this was going to be straightforward, it is not!”. Some of the routine challenges we face are:
- How do we experiment, when each experiment can cost us significant dollars?
- How do we detect that an identity is stolen - in real time?
- How do we underwrite users with short credit histories?
- How do we do all of the above in < 100 ms?
- What if analyses: how will a borrower behave if the market crashes?
- How do you predict consumer lifetime value when you don’t observe it?
THINGS YOU'LL DO
Build production models that underpin new and existing products. You own the end-to-end workflow of designing, developing, and deploying machine learning models. You coordinate with product managers, engineers, and even legal teams. Design experiments and analyze data. Your ad-hoc data analyses will decide which policies we adopt, where we expand our business, and which products we create.
WHAT PROBLEMS YOU'LL SOLVE
- Active Learning: How do you design a system where a manual review of transactions is expensive?
- Reject Inference: How do you learn about the performance of the population that the model rejects?
- Design of experiments: Can we continue to learn without losing our shirt?
- Conduct ad-hoc data analyses: your analyses will decide which policies we adopt, where we expand our business, and which products we create.
- Plan, get feedback, create, iterate, deploy. Own the DS workflow end-to-end, from idea to production.
WHAT WE LOOK FOR
- Degree in Statistics, Computer Science, Math, Physics, Econometrics, or a related quantitative field
- PhD with 1+ years industry experience, 3+ years with a MS, or 5+ years with a BS
- Solid understanding of and experience with machine learning fundamentals, such as gradient boosting, decision trees, regression, and unsupervised learning
- Strong programming skills in a scientific computing language such as Python. Experience using frameworks for machine learning and data science like scikit-learn, pandas, numpy
- Effectively communicate and present technical concepts and results to cross functional groups. Possess high quality writing ability.
- Nice to have: experience with big data technologies such as Spark, Hive, Hadoop, EMR, etc
- Persistence and patience and a great sense of responsibility - we create the decisioning that drives Affirm’s business!
If you got to this point, we hope you're feeling excited about the job description you just read. Even if you don't feel that you meet every single requirement, we still encourage you to apply. We're eager to meet people that believe in Affirm's mission and can contribute to our team in a variety of ways – not just candidates who check all the boxes.
At Affirm, "People Come First" is a core value and that’s why diversity and inclusion are vital to our priorities as an equal opportunity employer. You can learn more about our D&I efforts here.
We also consider qualified applicants with arrest and conviction records for positions in accordance with applicable laws, including the San Francisco Fair Chance Ordinance.