May 20, 2021

A new era of DevOps, powered by machine learning

While programming languages have evolved tremendously, at their core they all still have one major thing in common: having a computer accomplish a goal in the most efficient and error-free way possible. Modern languages have made development easier in many ways, but not a lot has changed in how we actually inspect the individual lines of code to make them error free. And even less has been done to keep your when it comes to improving code quality that improves performance and reduces operational cost. Where build and release schedules once slowed down the time it took developers to ship new features, the cloud has turbo charged this process by providing a step function increase in speed to build, test, and deploy code. New features are now delivered in hours (instead of months or years) and are in the hands of end users as soon as they are ready. Much of this is made possible through a new paradigm in how IT and software development teams collaboratively interact and build best practices: DevOps. Although DevOps technology has evolved dramatically over the last 5 years, it is still challenging. 

Productizing Machine Learning Models

Typically, there are three distinct but interconnected steps towards productizing an existing model: Serving the models; Writing the application’s business logic and serving it behind an API; and Building the user interface that interacts with the above APIs. Today, the first two steps require a combination of DevOps and back-end engineering skills (e.g. “Dockerizing” code, running a Kubernetes cluster if needed, standing up web services…). The last step—building out an interface with which end users can actually interact—requires front-end engineering skills. The range of skills necessary means that feedback loops are almost impossible to establish and that it takes too much time to get machine learning into usable products. Our team experienced this pain first-hand as data scientists and engineers; so, we built BaseTen. ... Oftentimes, serving models requires more than just calling it as an API. For instance, there may be pre- and/or post-processing steps, or business logic may need to be executed after the model is called. To do this, users can write Python code in BaseTen and it will be wrapped in an API and served—no need to worry about Kubernetes, Docker, and Flask. 

The timeline for quantum computing is getting shorter

Financial traders rely heavily on computer financial simulations for making buying and selling decisions. Specifically, “Monte Carlo” simulations are used to assess risk and simulate prices for a wide range of financial instruments. These simulations also can be used in corporate finance and for portfolio management. But in a digital world where other industries routinely leverage real-time data, financial traders are working with the digital equivalent of the Pony Express. That’s because Monte Carlo simulations involve such an insanely large number of complex calculations that they consume more time and computational resources than a 14-team, two-quarterback online fantasy football league with Superflex position. Consequently, financial calculations using Monte Carlo methods typically are made once a day. While that might be fine in the relatively tranquil bond market, traders trying to navigate more volatile markets are at a disadvantage because they must rely on old data. If only there were a way to accelerate Monte Carlo simulations for the benefit of our lamentably ladened financial traders! 

Pandemic tech use heightens consumer privacy fears

With user data the lifeblood of online platforms and digital brands, Marx said there were clear lessons for tech companies to learn in the post-pandemic world. Looking ahead, many study respondents agreed they would prefer to engage with brands that made it easier for them to control their data, up on previous years. Others called out “creepy” behaviour such as personalised offers or adverts that stalk people around the internet based on their browsing habits, and many also felt they wanted to see more evidence of appropriate data governance. Those organisations that can successfully adapt to meet these expectations might find they have a competitive advantage in years to come, suggested Marx. And consumers already appear to be sending them a message that the issue needs to be taken seriously, with over a third of respondents now rejecting website cookies or unsubscribing from mailing lists, and just under a third switching on incognito web browsing. Notably, in South Korea, many respondents said that having multiple online personas for different services was a good way to manage their privacy, raising concerns about data accuracy and the quality of insights that can be derived from it.

Why great leaders always make time for their people

When people can’t find you, they aren’t getting the information they need to do their job well. They waste time just trying to get your time. They may worry that, when they do find you, because you’re so busy, you’ll be brittle or angry. The whole organization may even be working around the assumption that you have no bandwidth. The sad truth, however, is that when you are unavailable, it’s also you who is not getting the message. You’re not picking up vital information, feedback, and early warning signs. You’re not hearing the diverse perspectives and eccentric ideas that only manifest in unpredictable, uncontrolled, or unscheduled situations—so, exactly those times you don’t have time for. And you’re not participating in the relaxed, social interactions that build connection and cohesion in your organization. So, though you may be busy doing lots of important stuff, your finger is off the pulse. But imagine being a leader who does have time, and how this freeing up of resources changes a leader’s influence on everyone below them. Great leaders know that being available actually saves time. A leader who has time would not use “busy” as an excuse. Indeed, you would take responsibility for time.

The road to successful change is lined with trade-offs

Leaders should borrow an important concept from the project management world: Go slow to go fast. There is often a rush to dive in at the beginning of a project, to start getting things done quickly and to feel a sense of accomplishment. This desire backfires when stakeholders are overlooked, plans are not validated, and critical conversations are ignored. Instead, project managers are advised to go slow — to do the work needed up front to develop momentum and gain speed later in the project. The same idea helps reframe notions about how to lead organizational change successfully. Instead of doing the conceptual work quickly and alone, leaders must slow down the initial planning stages, resist the temptation and endorphin rush of being a “heroic” leader solving the problem, and engage people in frank conversations about the trade-offs involved in change. This does not have to take long — even just a few days or weeks. The key is to build the capacity to think together and to get underlying assumptions out in the open. Leaders must do more than just get the conversation started. They also need to keep it going, often in the face of significant challenges.

Read more here ...