Feature
Artificial intelligence is transforming how we develop and manage technology, from automating code generation to optimizing infrastructure. But as AI-driven tools become more capable, the question arises: How much of your tech stack can be — or should be — built using AI?
The shift has the potential to accelerate development, reduce costs and provide wider access to advanced technology. However, it also raises concerns about long-term maintainability, security risks and reliance on black-box systems.
Acknowledging AI's Capabilities and Limitations
XData Group founder and CEO Roman Eloshvili describes AI as a tool to accelerate software development, but he emphasized that it does not inherently make programmers smarter or replace them — at least not yet. While AI excels at routine coding tasks and typo detection, it struggles with large-scale responsibilities like designing and managing software architecture.
AI can assist in building and maintaining core components of a tech stack, Eloshvili said, but its effectiveness depends on human oversight. The more skilled the developer using AI, the more advanced tasks it can handle. The main advantages of AI-driven development lie in speed and cost reduction, and Eloshvili predicts that AI will gradually take on more autonomous development tasks.
Regarding risks, Eloshvili believes that as long as proper review procedures are in place, there is little to worry about. Keeping AI-generated code readable and transparent ensures quality, and modern AI models like Claude Sonnet are already proficient in producing clear, functional code.
AI performs well across various domains, but Python and JavaScript are particularly well-suited for AI-driven development due to their widespread adoption and the abundance of training data. Companies integrating AI should recognize it as a tool, not a replacement for expertise. AI already operates at the level of a mid-level developer, making it more beneficial for businesses to invest in top-tier talent who can effectively leverage AI, rather than hiring junior developers.
AI Will Never Replace Engineers
Asked if AI would replace software engineers, Devansh Agarwal, a senior machine learning Engineer at AWS, gave a definitive no. According to him, large language models (LLMs) like ChatGPT and GitHub Copilot are powerful token predictors — excellent at generating text and code by predicting the next word in a sequence, but far from possessing human-like reasoning.
Agarwal said developers can integrate AI into the coding process in two primary ways: 1. direct prompting, where an engineer asks an AI to generate code and manually integrates it, and 2. IDE plugins like GitHub Copilot, which offer real-time code suggestions. While these tools improve efficiency, he warns of the security risk plugins can introduce, as they may expose an entire codebase to third-party providers.
Despite AI’s rapid advancements, Agarwal believes the fear of AI replacing software engineers fails to account for the work engineers do beyond writing code. An engineer working for any of the FAANG companies designs algorithms, debugs production issues, interacts with customers and make strategic decisions, tasks that even the largest AI models, with hundreds of billions of parameters, would struggle to handle. Even if AI scales to trillions of parameters, training such models would be environmentally unsustainable and economically impractical.
Instead, AI is best viewed as a productivity enhancer, automating repetitive tasks while leaving the critical thinking and problem-solving to humans. “I am ready to die on this hill,” Agarwal said. “As an engineer, I would be the happiest person in the world if AI could take over the boring parts of my job.”
Strengths and Limitations of AI in the Tech Stack
One area where AI shines is in generating boilerplate code, the repetitive, standardized code used for tasks like API integrations. In those cases, AI can save hours of manual effort by producing functional templates in seconds. However, when it comes to more complex, stack-building tasks, such as writing intricate unit tests or handling advanced debugging scenarios, AI often falls short, producing code that requires heavy modifications. To prevent AI-generated bugs from slipping into production, Agarwal suggests a few best practices:
- Compile and build AI-generated code to catch early errors.
- Ensure strong test coverage, particularly with unit and integration tests.
- Mark AI-generated code during review to help engineers scrutinize it more carefully.
- Use one AI model to verify another, as different models may highlight overlooked issues.
Overreliance on AI brings other risks too. Security remains a major concern. If AI misconfigures cloud infrastructure, for example, sensitive data could be exposed. Additionally, AI-generated code can introduce long-term maintainability issues, making debugging harder and locking teams into rigid, inefficient coding styles.
The Combination of GenAI and No-Code Tools
That said, Swimm co-founder and CTO Omer Rosenbaum highlights how AI is changing software development, particularly in the no-code and low-code space. The integration of LLMs into platforms like Bubble and WeWeb make application development more accessible to non-technical users, reducing reliance on traditional coding expertise. AI-powered no-code tools support this, he said, by:
- Generating functional app components based on user prompts.
- Automating business logic and workflows without custom scripting.
- Enhancing UI/UX design through AI-driven front-end generation.
According to Rosenbaum, the combination of AI and no-code democratizes software creation, allowing business teams and founders to go from idea to MVP faster and maintain their own products.
Looking ahead, AI-driven no-code tools will lead to:
- More accessible tech development, reducing the need for large engineering teams.
- Faster prototyping and iteration, enabling rapid experimentation.
- Evolving developer roles, with engineers focusing more on optimizing AI-generated workflows rather than building from scratch.
Ultimately, Rosenbaum shares Agarwal's belief that AI is not replacing engineers. Instead, it’s acting as a democratizing force, bridging the gap between coders and product users. Traditionally, there’s been a disconnect between the engineers who build the software and the end users who truly understand the business needs. AI eliminates that gap by allowing non-technical teams to take control of their own software while still benefiting from AI-assisted development.
Webinar
Sep
25
CMS Briefing: A Live Look at What’s Next in AI-Driven Platforms
Learn how leading organizations are using AI‑driven tools to publish faster, personalize smarter and stay secure.
Register
Webinar
On demand
Ready or Not: How Data-First Organizations Are Unlocking Agentforce Potential
Learn how to cut through the noise, activate Agentforce and build a Salesforce AI strategy that actually delivers.
Watch Now
Webinar
On demand
AI in Customer Service: Faster Resolutions, Happier Customers
Don’t let rising demand burn out your team. See how to build a smarter, more resilient support org.
Watch Now
Webinar
On demand
From Hype to High-Impact CX Strategies That Actually Scale
Turn buzzworthy AI and outsourcing trends into measurable CX wins with fresh 2025 data.
Watch Now
Webinar
On demand
Insights to Action Rethinking the Contact Center for Real Business Impact
Join our exclusive webinar to hear CX executives share their innovative strategies for transforming service delivery.
Watch Now
Webinar
On demand
Accelerating Healthcare Ops with AI
Sign up to see how your peers are putting AI to work — and making it stick.
Watch Now
Looking ahead, Rosenbaum predicts the impact of no-code AI will be undeniable. More people will be able to build tech, reducing dependence on expensive engineering teams. Prototyping will accelerate, making iteration and innovation faster than ever. And while developers won’t disappear, their roles will evolve—instead of writing everything from scratch, they’ll focus on optimizing AI-generated workflows.
For Rosenbaum, the takeaway is clear: AI isn’t here to replace developers. It’s here to empower more people to create technology. Businesses that embrace this shift will be the ones that launch and scale products at unprecedented speed.