Applied Deep Learning, Graduate Intern (Master's or PhD)

Company Description

It all started with an idea at Block in 2013. Initially built to take the pain out of peer-to-peer payments, Cash App has gone from a simple product with a single purpose to a dynamic ecosystem, developing unique financial products, including Afterpay/Clearpay, to provide a better way to send, spend, invest, borrow and save to our 47 million monthly active customers. We want to redefine the world’s relationship with money to make it more relatable, instantly available, and universally accessible.

Today, Cash App has thousands of employees working globally across office and remote locations, with a culture geared toward innovation, collaboration and impact. We’ve been a distributed team since day one, and many of our roles can be done remotely from the countries where Cash App operates. No matter the location, we tailor our experience to ensure our employees are creative, productive, and happy.

Check out our locations, benefits, and more at cash.app/careers.

Job Description

This opportunity is only open to students who are currently enrolled in either a Masters or PhD program and offers an 8-month term (with a possibility for extension) on Cash App's Risk AI team starting in Summer or Fall of 2024. If you've already graduated (or are set to graduate before May 2023) please apply to one of our full-time roles.

Machine Learning is an integral part of how we design products, operate, and pursue Cash App's mission to serve the unbanked as well as disrupt traditional financial institutions. Our massive scale and deep trove of transaction data create an endless number of opportunities to use artificial intelligence to better understand our customers and offer new products and experiences that can improve their lives. We are a highly creative group that prefers to solve problems from first principles; we move quickly, make incremental changes, and deploy to production every day.

The objective of this project is to systematically test and validate advanced sequential models (including bi-directional RNN, auto-regressive CNN, transformers, and neural controlled differential equations) on time-series data generated within Cash. Cash App collects a variety of signals about how our customers interact with the platform - we will determine the best way to make use of these data using state-of-the-art deep learning methods.

Deep sequential models offer the potential to significantly improve our ability to detect fraudulent behaviour while reducing the time-consuming (and hence expensive) process of manual feature-creation and curation. Models will be benchmarked for automated extraction and detection of fraud patterns within sequential transactional data (e.g. cash in, cash out, p2p, etc).

Over the course of your term, you will:

• Craft standardized and reusable datasets for consistent benchmarking and testing of sequence models
• Provide quantitative assessment of the advantages and disadvantages of different model classes based on accuracy, precision/recall, training cost, training time, data requirements, and interpretability
• Create a modular and well documented codebase which can be easily used for other sources of sequential data within the Block ecosystem Technologies we use (and teach):

Technologies we use (and teach):

• Python, NumPy, Pandas, PyTorch, TensorFlow, keras, JAX, Julia
• MySQL, Snowflake, GCP/AWS and Tableau
• Java

Qualifications

You have:

• 1-2 years of experience with applied Deep Learning
• Proven ability to implement in practice neural network architectures described in literature using frameworks such as PyTorch or Tensorflow
• Experience with advanced techniques like graph embeddings, irregular sequence modeling and time-series forecasting, uncertainty quantification, anomaly detection, and neural ODEs is a big plus
• An appreciation for the connection between the software and models you build and the experience it delivers to customers
• Have a curious, passionate, growth-oriented mindset

Additional Information

Block takes a market-based approach to pay, and pay may vary depending on your location. U.S. locations are categorized into one of four zones based on a cost of labor index for that geographic area. The successful candidate’s starting pay will be determined based on the candidate's work location and may be may be modified in the future.

Zone A: USD $51.00
Zone B: USD $48.45
Zone C: USD $45.90
Zone D: USD $43.35

To find a location’s zone designation, please refer to this resource. If a location of interest is not listed, please speak with a recruiter for additional information. 

To find a location’s zone desig