Upcoming Events
 

ML Design Patterns and Designing ML Infrastructure
February 24 | 6:00 PM | Online

Design patterns are formalized best practices to solve common problems when designing a software system. As machine learning moves from being a research discipline to a software one, it is useful to catalog tried-and-proven methods to help engineers tackle frequently occurring problems that crop up during the ML process. In this talk, I will cover five patterns (Workflow Pipelines, Transform, Multimodal Input, Feature Store, Cascade) that are useful in the context of adding flexibility, resilience and reproducibility to ML in production. For data scientists and ML engineers, these patterns provide a way to apply hard-won knowledge from hundreds of ML experts to your own projects.
 

Data Science Product Management
March 20 | 12:00 PM | Online

When put into service solving customer needs, data science can be a critical differentiator for digital products in ever-more-competitive markets. But productizing data science presents a unique set of challenges, and often leaves product managers and data scientists struggling to find common ground and a shared language. In this talk, product coach and consultant Matt LeMay shares the lessons he's learned building bridges between product management and data science at companies like Bitly, Songza, and Spotify. Expect a candid, direct, and entertaining conversation about mistakes made, lessons learned, and suggestions for how to move forward.
 

Pitch! - Yourself, Your Company, Your Project, Your Ideas!
April 3 | 1:00 PM | Online

Looking for a job? Looking to hire someone? Trying to get your project started? Grab your best pitch and come share with the Data Works MD community. We need speakers! If you would like to speak, please register here.

Past Events

Opportunities

Data News and Articles

Why Is It So Hard to Become a Data-Driven Company? — To compete today, companies need to be data-driven. But for mainstream, legacy companies, that’s easier said than done. Despite a decade of investment and the adoption of Chief Data Officers, this survey of Fortune 1000 senior executives finds that many companies are still struggling against not just legacy tech, but embedded cultures that are resistant to new ways of doing things — over 90 percent of companies surveyed reported culture was their biggest barrier. In response to this, leaders should do three things: 1) focus their data initiatives on clearly identified high-impact use cases, 2) reconsider how their organizations handle data, and 3) remember that this transformation is a long-term process that requires patience, fortitude, and focus. Tags: Data, Company

Managing Data as a Data Engineer — As we have more data and as data grows larger, managing data becomes quite a challenge. In this article, the author shares some of the things that learnt while managing terabytes of data in a fintech company. Tags: Data

How Spotify Optimized the Largest Dataflow Job Ever for Wrapped 2020 — This article discusses how Spotify optimized and sped up elements of their largest Dataflow job using a technique called Sort Merge Bucket (SMB) join. Tags: Big Data

Why Some Models Leak Data — Machine learning models use large amounts of data, some of which can be sensitive. If they're not trained correctly, sometimes that data is inadvertently revealed. Tags: Data

Interview With an Applied Scientist at Amazon — What does an average day look like at Amazon? A quick interview about working at Amazon. Tags: Career

Google Research: Looking Back at 2020, and Forward to 2021 — Extensive article on research areas of 2020 including COVID-19, responsible AI, reinforcement learning, robotics, and quantum computing. Tags: Research

(At Least) 5 Ways Data Analysis Improves Product Development — At the company Mode, they believe that iterative analytics is a critical method to how you answer any meaningful product question. Traditional BI tools are great at answering static questions (examples: how many purple socks did we sell today?), but not the big-picture questions that direct decisions on what big swings to take in your product roadmap. This article describes five ways that they see data science as necessary to prioritize the best product development choices. Tags: Team, Product Development

Cultivating Algorithms: How We Grow Data Science at Stitch Fix — Data, by itself, is seemingly chaotic and unorganized. Yet, when properly processed, it can provide a competitive advantage or even enable completely new services. This article covers how Stitch Fix discovers innovative capabilities through curiosity-driven tinkering. Tags: Team

Coronavirus, a Visual Rundown — A quick rundown of the useful visuals about COVID-19. Tags: COVID, Visualization

Ditching Excel for Python – Lessons Learned from a Legacy Industry — Since 2017, the author has observed a radical shift in data analysis methodologies. Excel-based models, which had seemed top-of-the-line suddenly were too slow and too rigid; Integration with 3rd party data sources, which was once a luxury, became the norm; And analysts began to utilize scripts to accomplish many labor-intensive tasks typically performed by hand or in spreadsheets. Enabling this change is a suite of accessible Python-powered tools. Tags: Python

Machine Learning is Going Real-time — There seems to be little consensus on what real-time ML means, and there hasn’t been a lot of in-depth discussion on how it’s done in the industry. This article discusses two levels of real-time ML: making predictions in real-time and incorporating new data and updating models in real-time. Tags: ML
 

How-To's and Tutorials

Data Tools and Resources