AI coding tools are now part of many software teams, but trust in those tools still lags behind their growing use. Developers rely on AI to help write code and speed up tasks, yet questions remain about reliability, accountability, and how to measure real gains.
For companies that build large digital platforms, the challenge is not simply adopting AI tools. It is deciding how those tools should fit into existing engineering practices without weakening quality or oversight.
AI News spoke with Idan Zalzberg, Chief Technology Officer at Agoda, about how the travel platform is approaching this shift. His comments show how one large technology company is testing where AI helps engineers move faster and where human review still matters.
Engineers remain accountable for AI-generated code
One of the clearest decisions Agoda has made is that AI does not change who is responsible for the code that enters production.
“First, it’s critical for us to maintain that engineers are accountable and responsible for their work, whether they used AI or not,” Zalzberg said. “AI is just a tool, and engineers within different teams should have the oversight and measures needed to ensure the code it generates is correct and maintainable.”
That principle shapes how Agoda treats AI-generated output in its development process. Code produced with the help of AI is still tied to the engineer who prompted the tool.
“As such, code generated by AI is still attributed to the specific engineer who used AI to produce it,” he said.
This approach avoids a common concern around AI-assisted coding: the risk that automation might weaken accountability in engineering teams. At Agoda, the responsibility for the code remains unchanged even if the method of writing it shifts.
Existing engineering safeguards still apply
AI tools may speed up coding tasks, but Agoda has not created a separate process to validate AI-generated code. Instead, the company runs it through the same checks used for any other software change.
“In terms of practices to ensure AI-generated code is high quality, we rely on the same mechanisms we use for human-written code,” Zalzberg explained.
Those mechanisms include linters, static analysis tools, automated testing, and gradual rollouts. These safeguards help detect errors before software reaches users.
“Some of these processes are themselves assisted by AI, which helps us apply them at larger scale regardless of how the code was written,” he said.
For engineering teams, this approach keeps the core development workflow stable while allowing AI tools to operate inside it.
Reliability remains a challenge for AI coding tools
Despite rapid progress in AI models, reliability remains a major concern for developers. Zalzberg said the unpredictability of generative systems changes how engineers must evaluate results.
“The biggest challenge when using AI is its unpredictability,” he said. “Unlike ‘standard’ software, we can’t always tell exactly what the code will do by simply reading it.”
The difficulty extends beyond code generation. AI systems that produce text or suggestions may not return the same result every time they run.
“We can’t assume running the same code multiple times will produce identical results,” Zalzberg said.
Because of this, Agoda engineers focus heavily on evaluation methods designed for AI systems. These approaches measure performance across many outputs rather than relying on a single test result.
“The way to address this is by becoming experts in evaluation, which is a critical skill for AI engineers,” he said.
Engineers are shifting toward supervision and decision-making
AI tools are changing how developers interact with software projects, but Zalzberg argues that the core role of engineers has not changed.
“It’s true that engineers no longer write all the code themselves and instead often instruct an AI assistant to generate parts of it,” he said. “However, that was never the core role of an engineer.”
Instead, engineers remain responsible for making technical decisions that shape a system. “The main responsibility has always been making technical decisions at different levels to address business problems,” Zalzberg said.
This includes choosing the architecture, selecting technologies, and defining data structures. AI may assist with writing pieces of code, but the broader design still requires human judgment.
“When AI produces code, engineers still need to challenge it and ensure the output aligns with the technical direction of the project,” he said.
That shift also means engineers must develop stronger questioning and communication skills when working with AI systems. “Because they interact with the underlying code differently, they need to be excellent at asking questions and using curiosity to ensure the output is correct,” Zalzberg added.
Measuring productivity gains from AI coding tools
Many developers report saving time when using AI coding tools, but translating those gains into broader engineering outcomes can be difficult.
Agoda saw early signs of improvement when it first introduced AI tools. “When we first rolled out AI tools, we were able to run controlled experiments that showed roughly a 27% productivity uplift,” Zalzberg said.
As AI usage expanded across teams, however, those clean comparisons became harder to maintain. “Today, AI usage is so widespread that we no longer have a clean benchmark for comparison,” he said.
Instead of measuring individual productivity, Agoda now watches broader indicators such as how quickly code changes move through the development process. “We track our overall velocity improvements over time, such as pull request throughput,” Zalzberg said.
Interestingly, the benefits did not appear immediately.
“In many areas it took a considerable amount of time before improvements became visible,” he said. Engineers needed time to integrate AI tools into their workflows and existing codebases.
Building governance for AI across the organisation
Beyond engineering productivity, Zalzberg believes mature AI use requires clear governance around how models are used inside a company.
At Agoda, one step toward that goal has been creating a central gateway that connects teams to different AI models. “We developed a single AI gateway for all models so we can monitor usage centrally,” he said.
The gateway helps track how teams use AI systems and how much they cost to run. It also gives the company visibility into where AI is being applied.
Another priority is deciding which decisions should remain under human control. “AI-driven decisions should either be reviewed by humans in some form or limited to areas where mistakes have no material consequences,” Zalzberg said.
For example, AI might help rank hotel recommendations but should not directly set the price customers pay for a room. “Being intentional about what AI controls — and what it does not — is critical,” he said.
AI coding adoption remains a work in progress
AI tools are already shaping how engineers work, but companies are still learning how to manage them safely and effectively.
At Agoda, that process involves experimentation alongside careful oversight. The company continues to watch how AI tools perform and how engineers adapt to them.
As Zalzberg put it, trust in AI grows gradually as teams learn where the technology works well and where it still requires caution.
“Trust is something that needs to be earned,” he said.
(Photo by Agoda)
See also: AI coding assistants may influence which languages developers use
Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and is co-located with other leading technology events, click here for more information.
AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here.