top of page

Devin, English Set to Be the Most Popular Programming Language

Devin AI, a new prospect in the software industry, is causing quite a stir (not to be mistaken with Devin Townsend). At the same time Nvidia's CEO sees a future where everyone has the potential to be a programmer. Are we seeing a dramatic shift in the way software is being developed?

Human vs AI combat in software engineering

Generative AI and RAG - the future of programming

Artificial Intelligence (AI) is gradually penetrating every aspect of our economy, from farming to healthcare, and even transportation. Nvidia's CEO, Jensen Huang, however, takes this further by picturing a reality where AI is not just an auxiliary tool, but an active collaborator. He's optimistic that we're approaching a time when anyone will be able to program, regardless of their background or expertise level. This progressive view is becoming viable due to advances in generative AI. Generative AI or large language models in general can comprehend various forms of input such as speech and transform them into computer programs. Thus, English is on track to be the most widely used programming language.

Considering my experience in the EdTech and MarTech sectors, I warmly welcome transformational developments shaping the tech future. I find Huang's foresight not just fascinating but invigorating considering the potential influence it holds over SaaS businesses, startups, and the IT market at large.

Generative AI heralds a future where programming is inevitable, thanks to its ability to perceive and respond to a variety of inputs. So, how does generative AI operate? It employs machine learning algorithms that generate top-tier content from minimal input, learning from data patterns to produce similar data. Be it crafting a musical composition in Mozart's style or painting in the manner of Van Gogh, generative AI offers endless possibilities. Now this technology is venturing into the domain of programming.

What is Devin AI?

Recently CognitionLabs, a New York-based firm introduced Devin, the AI software engineer who can solve almost one in every seven programming issues on StackOverflow — a widely used informational platform for programmers. This accomplishment surpasses any existing model, marking a turning point in the coding landscape. With generative AI, the programming community can look forward to AI platforms that can manage more intricate tasks and code independently. Allegedly it can complete complex software engineering tasks.

What can Devin AI do?

Devin AI can:

  • undertake complex engineering tasks,

  • make design choices,

  • remember the relevant context and background data at every stage,

  • learn over time, and correct its errors,

  • find bugs in codebases,

  • it can use regular developer tools including the shell, code editor, and browser within a computing environment, much like a human software engineers,

  • additionally, Devin can actively partner with you, providing real-time progress updates and feedback.

Upon its effect on SaaS companies, tools like Devin AI can potentially alleviate a global programmer deficit and save them from monotonous tasks. These tools could not only speed up software production but could also cut labor costs and enhance software quality. However, the introduction of generative AI may require continuous investment in training and policy adjustments, which is always easier said than done.

While generative AI in programming offers promising prospects, it does pose its own set of risks and difficulties. While Devin AI can create code, understanding, debugging, and requirement stating tasks are still fundamentally human engineers tasks. There are also valid concerns about AI-generated code’s security, privacy, and legal aspects. Additionally, generative AI might further alienate professionals without access to this technology or accompanying training.

Will Devin AI replace coders?

Congenial AI software engineer at work

In essence, generative AI holds the potential to revolutionise programming, with Devin standing as proof. Huang's vision could lead us into an age where anyone can play a role in programming, leading to a more inclusive technological era.

However, this path presents challenges, including overcoming accessibility and training hurdles, along with addressing security and privacy issues. As we progress through this new era, it will be up to businesses, governments, and society as a whole to tackle these challenges and leverage the opportunities that generative AI offers.

Regardless of whether we end up in a world where 'everyone is a programmer', the impact of generative AI on programming's future is likely to be substantial.


bottom of page