Data Science and Databases

Unwanted AI bias is already a widespread problem. Machine learning models can replicate or exacerbate existing biases, often in ways that are not detected until release. So what can be done about it?

We reduce boilerplate with common OOP patterns, but often, the code overhead of using patterns is almost the same—if not bigger—than if we’d used boilerplate code in the first place.

In aspect-oriented programming tutorial, we implement two functions (aspects) that allow for generic database connection management and caching. These allow us to mark methods that, when called, will automatically have the required functionally present in through IL weaving.

Migrating in either direction between Microsoft SQL Server and Oracle Database? There are some significant pitfalls to be aware of…

Data collection and preparation slow down traditional NLP projects. However, transfer learning and BERT can reduce the amount of data required and change the way companies execute NLP projects.

Nearly every app requires an authorization system on top of an authentication system. We often need a fine-grained permissions model to allow only specific users to access certain resources while simultaneously restricting access for everyone else.

In this tutorial, Toptal Freelance Firebase Developer Joaquin Cid teaches how to build just such a system. Called role-based authentication, we’ll learn how to implement this system in an API using Node.js and Firebase.

Subscription-based apps need a special approach when it comes to analytics. Learn how a variation on the traditional sales funnel helps you work with LTV, subscriber retention, and churn rates to make your app better—and more profitable.

SQL window functions extend aggregate logic by operating across rows without grouping them. Learn how they work, when to use them, and how to apply them in real-world queries.

What do you do when you come across a dataset that’s simply too big to allow fast queries? Google BigQuery could be part of the solution, though it’s no silver bullet.

In this article, Toptal Java Developer Maksym Goroshkevych explains how BigQuery can drastically improve query performance, and when you should and should not use it.

The increasing accuracy of machine learning systems has resulted in a flood of applications using them. As machine learning models matured and improved, so did ways of attacking them.

In this article, Toptal Python Developer Pau Labarta Bajo examines the world of adversarial machine learning, explains how ML models can be attacked, and what you can do to safeguard them against attack.

Building a mobile app isn’t cheap—and the harsh reality is that most apps fail. Find out how to leverage fundamental analytics metrics to make sure your app actually turns a profit.

Static websites are simple and fast, but it’s also possible to automate updates to them to keep them fresh and relevant. Find out how to generate daily-updated HTML pages that track the latest releases of popular GitHub repositories.

You can just write sequentially executed code in JavaScript, but should you?

In this article, Toptal Freelance JavaScript Developer Omar Waleed tests the widespread Node.js belief that synchronous code degrades performance and is, in a sense, just plain evil. Is this really true?

Automatically scaling container deployments in a microservices-based app architecture is downright luxurious…once it’s set up. But what’s the best way to tune an app’s orchestration parameters?

Data warehouses aren’t exactly a new concept, but industry demand for data science services, coupled with the rise of AI and machine learning, is making them more relevant than ever.

In this post, Toptal Data Warehouse Developer Chamitha Wanaguru outlines three basic principles you need to keep in mind when developing a new data warehouse.

Without proper state management, your Angular app will become a UX nightmare. But even with that solved, persisting state across sessions and devices can be tricky. This tutorial shows how to leapfrog that challenge using Firebase.

The whole point of smart contracts is that they need to be more secure and efficient than traditional contracts. So where do smart contract oracles fit in?

In the final installment of our three-part series, Toptal Blockchain Developer John R. Kosinski explains the role of oracles in the evolution of trust.