Will AI Finally Make TDD Practical? | Diffblue’s Animesh Mishra
Mar 18, 2025
auto_awesome
Animesh Mishra, a Senior Solutions Engineer at Diffblue, joins the conversation to explore the unfulfilled promise of Test Driven Development (TDD). He discusses how deterministic AI can bridge the gap between theory and practice, potentially writing two years' worth of tests for legacy code in just two hours. Animesh also highlights the importance of building trust in AI for testing, emphasizing how it can alleviate developers' cognitive strain while improving software quality and efficiency.
AI has the potential to revolutionize Test Driven Development by automating testing processes, significantly reducing time and enhancing productivity.
The challenges of implementing effective TDD highlight the ongoing tension between developers' enthusiasm and the practical limitations they face in execution.
Deep dives
The Sandwich Generation in Tech
The sandwich generation refers to individuals who simultaneously care for both their children and aging parents, a situation increasingly common in the tech workforce. Approximately 11 million people in the U.S. fall into this demographic, with millennials and Gen Z constituting about a third. These individuals often face significant lifestyle and financial sacrifices, with many reporting impacts on their own financial security due to caretaking responsibilities. The flexibility of remote work has been noted as a potential alleviator, allowing caretakers to maintain their professional duties while being more present for their families.
Privacy Concerns and Google Chrome
Recent developments in Google Chrome's handling of user privacy have sparked concerns, particularly with the removal of popular ad-blocking extensions like uBlock Origin. This shift is seen as part of a broader trend where Google is exerting pressure to define standards beneficial to its advertising model, likened to a modern-day AOL. Many users are uncomfortable with the implications of such changes, as they potentially compromise the autonomy and privacy of internet browsing. Alternative solutions, such as using DNS services like Piehole, have gained popularity among users seeking to maintain their privacy.
AI in Hiring Processes
A recent incident has highlighted the potential risks of AI in hiring, involving candidates who utilized deepfake technology to impersonate others during interviews. This case serves as a cautionary tale, emphasizing the need for robust verification methods to confirm the identities of applicants. As AI continues to influence the hiring landscape, organizations must develop strategies to differentiate between genuine candidates and AI-assisted impersonators. The importance of maintaining integrity in the hiring process while embracing new technologies is underscored by the evolving challenges companies face.
Transforming Test-Driven Development with AI
The conversation around test-driven development (TDD) reveals a paradox: while developers are enthusiastic about the concept, many struggle to implement it effectively. AI is being positioned as a solution to streamline the TDD process, potentially saving developers time and improving consistency in testing. By automating mundane aspects of testing, AI could allow developers to focus on more impactful tasks, thus enhancing productivity. However, as AI tools proliferate, the industry is challenged to integrate these technologies without sacrificing quality or introducing complexities in debugging.
The promise of Test Driven Development (or TDD) remains unfulfilled. Like many other forms of aspirational development, the practice has fallen victim to countless buzzword cycles. What if the answer is already in our toolbox?
This week, host Andrew Zigler sits down with Animesh Mishra, Senior Solutions Engineer at Diffblue, to unpack the gap between TDD's theoretical appeal and its practical challenges.
Animesh draws from his extensive experience to explain how deterministic AI can address the key challenges of building trust in AI for testing. These aren’t LLMs of today, but foundational machine learning models that can evaluate all possible branches of a piece of code to write test coverage for it. Imagine writing two years worth of tests for a legacy codebase… in two hours… with no errors!
If you enjoyed this conversation about the gaps between theory and execution in engineering culture, be sure to check out last week's chat with David Mytton about shift left adoption by engineering teams.