This article is based on a February 2017 report originally published by The Hindu BusinessLine (thehindubusinessline.com), covering statements made by senior technology executives at Mphasis, KPMG, and Capgemini.
The Provocative Claim: “Coding Is Dead”
In February 2017, Nitin Rakesh, then newly appointed CEO of IT services firm Mphasis, made a statement that ricocheted around the industry: “Coding is dead. I don’t need to do any native development anymore.” Speaking to BusinessLine, Rakesh argued that large public cloud platforms — Amazon Web Services (AWS), Salesforce, Microsoft Azure, and IBM Watson — had fundamentally changed how software is built and delivered.
The claim was deliberately provocative, but the underlying point deserves careful examination rather than a knee-jerk reaction. What Rakesh and other executives were describing was not the end of programming as a discipline, but a structural shift in how software is assembled and delivered at enterprise scale.
What “Cloud First, Mobile First” Actually Means for Engineers
Rakesh’s argument centered on a straightforward observation: banks and enterprises that once wrote custom code in native Java on Unix machines are increasingly running workloads on managed cloud platforms. On those platforms, much of the heavy lifting — infrastructure provisioning, runtime management, security patching — is handled by the provider.
The practical consequence for engineers is a shift in the value chain:
- Less time writing low-level boilerplate. Billions of lines of open-source code are freely available and can be composed — much like assembling Lego bricks — rather than written from scratch.
- More time on configuration and integration. As Rakesh noted, frameworks such as Node.js allow engineers to orchestrate existing routines rather than write native implementations from scratch.
- Faster time to market. Reusing battle-tested components reduces development cycles and shifts competitive advantage toward teams that can integrate quickly and iterate rapidly.
The KPMG Counterpoint: Programming Knowledge Still Matters
Not everyone accepted the “coding is dead” framing without qualification. Akhilesh Tuteja, Partner and National Head for Technology and BPM at KPMG, offered a more nuanced view:
“It is true the way we program today may not exist in the near future, as there’ll be more of reusable code that’ll be in play. However, you will require some level of programming knowledge to understand how to put these blocks together and make sense of them.”
Akhilesh Tuteja, Partner and National Head (Technology and BPM), KPMG
This is the more defensible position for working engineers. Composing cloud-native services, configuring infrastructure-as-code pipelines, writing Terraform modules, or building serverless functions all require genuine programming literacy. The abstraction layer has moved up, but it has not disappeared.
Automation Accelerates the Shift
Platforms such as AWS, Azure, and IBM Watson do not just provide managed infrastructure — they also provide tooling that automates testing, deployment, and monitoring. What once required days of manual testing can be triggered with a single pipeline step. This is the DevOps proposition in practical terms:
- CI/CD pipelines automate build, test, and release cycles.
- Infrastructure as Code (IaC) tools like Terraform and AWS CloudFormation reduce manual provisioning to configuration files.
- Managed services (databases, queues, identity, ML inference) replace large categories of custom code with API calls.
- Low-code and no-code platforms extend development capabilities to non-specialists for well-defined problem domains.
The cumulative effect is that a small, skilled team using modern cloud tooling can deliver what previously required large groups of native developers — not because coding is irrelevant, but because automation has dramatically raised the productivity ceiling per engineer.
Industry-Wide Re-Skilling: What IT Services Firms Were Doing
By 2017, major Indian IT services firms — TCS, Infosys, Wipro, and Capgemini — were already investing heavily in re-skilling large portions of their workforces. The focus areas reported at the time included:
- Agile and Scrum methodologies
- Design thinking and rapid prototyping
- Cloud platform certification (AWS, Azure, GCP)
- Reusable-code and component-driven development
Patrick Nicolet, Group Executive Board member at Capgemini, was quoted summarizing the urgency: “If you look at delivery in digital, it is about design thinking, agile, rapid prototyping and everything that is fast and furious. The way we have designed our delivery so far is not fast and furious.”
What This Means for DevOps Engineers Today
The 2017 conversation anticipated trends that have only accelerated. For DevOps and platform engineers, the practical takeaways are:
- Understand platforms deeply, not just syntax. AWS, Azure, and GCP are now primary engineering environments. Knowing how to wire services together securely and efficiently is a core skill.
- Programming literacy remains non-negotiable. Writing Bash, Python, Go, or HCL (Terraform) is still essential — the abstraction layer has moved, not vanished.
- Automation is the multiplier. Engineers who can build and maintain robust CI/CD pipelines, automated testing suites, and infrastructure pipelines consistently outperform those who cannot.
- Reusability is an architectural discipline. Internal developer platforms, shared Terraform modules, and standardized container images are the modern equivalent of Rakesh’s Lego-block analogy.
- Speed is the competitive variable. “Fast and furious” delivery — iterating quickly based on real user feedback — is increasingly what separates effective engineering organizations from slow ones.
The Verdict: Transformation, Not Death
“Coding is dead” was always a headline engineered for attention. The reality is more nuanced: the nature of software engineering is transforming. Low-level native development has a shrinking role at the application layer, while platform engineering, infrastructure automation, and systems thinking have a growing one.
Developers who adapt — learning cloud platforms, embracing automation, and mastering reusable-component architectures — are not facing obsolescence. They are becoming significantly more productive. Those who treat any single language or framework as a permanent career foundation, without adapting to platform-centric delivery, face a harder road.
The most durable skill, as it turns out, is the ability to learn — which is precisely what these executives were arguing in 2017.