BY Fast Company Contributor 4 MINUTE READ

A black and white collage swirls on a hospital computer screen: an MRI scan. Even to the trained eye, it’s difficult to see anything amiss. But an AI-powered computer program—using software trained on thousands of similar images—sees the early signs of cancer. Detected early, the prognosis for recovery is remarkably good.

Behind this amazing innovation, like so many others in our lives today, is something both omnipresent and easily overlooked: code.

Whether it’s health screenings or the way we order food, do our banking, or drive cars, the biggest innovations that shape our lives today are powered by software—specifically the roughly 30 million developers worldwide who write the code that creates it. How fast we progress, in business and society, comes down to how fast, and how efficiently, they can work.

But the efficiency of converting raw ideas to code and, ultimately, shipping it to end users is anything but assured. In fact, this “innovation supply chain,” much like our physical supply chains, faces serious pinch points. Today, developers—the driving force behind so much tech advancement—are being held back by outdated processes and a lack of tools. Clearing those bottlenecks is much more than a matter of convenience: I believe it has a major impact on the pace at which humanity can progress.


We saw during the pandemic how something as simple as a lack of shipping containers can impact commerce and productivity globally. Those same types of backups occur in the innovation supply chain, only the bottlenecks that occur in the production of software aren’t so much physical; rather, they are process-oriented.

A huge issue holding developers back right now is the amount of toil that creeps into their daily work. As researcher Vivek Rau identified in his work with Google, toil refers to any process that is “manual, repetitive, automatable, tactical, devoid of enduring value, and that scales linearly as a service grows.” In layperson’s terms, toil is all the administrative and busy work that goes along with writing software.

This may sound merely annoying, but as software has become more complex, so has the amount of toil developers encounter—taking precious time away from the creative work that has the potential to change our lives.

I’ve seen this impact firsthand. When I started as a developer, the best part of my day was putting on my headphones, pumping some music, and getting into the zone to write some code. But throughout my career, that has become a smaller and smaller part of a dev’s job. Software engineers now spend less than half their days writing code, with some estimates putting that number as low as 20%. The rest of their time is bogged down by tasks such as testing and shipping code, waiting for builds to get done, or administrative roadblocks like needing approval to go forward with the next stage of a code change.

This has consequences on multiple levels: for developers, it’s extremely frustrating and stifles creativity and enjoyment on the job—even leading to burnout and contributing to developer shortages. For businesses racing to innovate, it can slow output and limit the ability to bring products to market in a timely manner. For the rest of us, it translates to a slowed pace of life-changing innovation. I have found that part of the reason advancements like self-driving cars or personalized medicine aren’t commonly accessible right now is surprisingly mundane: developers are being held back by mountains of toil.


Unblocking the innovation supply chain requires bringing developer experience into the spotlight. User experience and even employee experience have become a common focus of organizations looking to improve output, but the conditions developers work under—and what they’re asked to spend time on—have long been neglected. Thankfully, that’s starting to change.

The concept of “DX” has emerged in tech circles in the last few years and is now starting to make its way into the wider business community. More companies are realizing that getting more out of their developers means making room for creativity and reducing the complexity of their workflow.

Using technology to automate rote processes, like testing, security, and delivery, is one way to do just that—and it addresses a deep irony that lingers in the developer space. For all the cutting-edge AI and automated tools software engineers create for other industries, tools to help developers in their own work lag far behind.

I noticed this early on in my career as a developer and decided to do something about it: starting a series of companies to equip developers with a more robust, reliable toolkit. The latest iteration of these platforms, incorporating AI and machine learning, has helped companies drastically cut down on toil and free up developers’ time for more important work in fields as diverse as finance, transportation, and health care.

A final issue is one of company culture: Developers work best when empowered with the trust and autonomy to do great work. But too many managers and organizations bog down their teams with arbitrary processes or administrative tasks. Even simple changes are held up by required approvals. A far better approach is to entrust developers with freedom to execute within defined guardrails. This delicate balance of autonomy and oversight allows for rapid innovation without compromising results. We do this with our team by setting parameters on cost, security, or UX, but otherwise giving them free rein.

The payoff for these changes goes beyond benefits to businesses. The next wave innovation goes far beyond consumer conveniences like media streaming or food delivery apps. More powerful batteries could help mitigate climate change, biometrically targeted treatments may eradicate some cancers, and 3D-printed organs may extend the lives of countless people.

Clearing toil from developers’ paths—and unblocking the innovation supply chain—means this future can become a reality sooner rather than later.


Jyoti Bansal is a multi-unicorn founder, entrepreneur and investor. He is the CEO of Harness, which uses AI to simplify software delivery.