Generative AI – Transforming software product engineering

By Bhavesh Ved, Chief Digital Officer, BDO Digital

In the dynamic realm of software product engineering (SPE), Generative AI (GenAI) has emerged as a catalyst for innovation. This innovative technology combines artificial intelligence with agile methodologies, reshaping how software products are conceived, developed, and refined. This article examines the benefits, challenges, and adoption strategies of AI, and how existing agile teams can harness its power to elevate their productivity and creativity.

Impact of GenAI in SPE Challenges in the adoption of GenAI in SPE
GenAI is accelerating the end-to-end cycle of software products from Ideation to Implementation and even Post-implementation.

GenAI tools are enabling rapid prototyping to bring new ‘Ideas to the Screen’ faster. Designers can unleash their creativity through a multitude of design alternatives.

Code Assists/ Copilots are fast becoming the ‘buddy’ for the developer community. AI is empowering developers to generate new code, debug code, test code, migrate code, optimise code, etc.

Data Assists/ Copilots with ‘Text to Insights’ functions are being used for data transformation, dynamic reports and dashboard generation.

QA Assists/ Copilots are leveraging the power of GenAI for test case generation, synthetic test data generation, test automation, etc.

One big benefit observed is code and API documentation of the legacy code.
GenAI’s power of personalisation makes systems more intuitive and creates a special bond with end users.
 

AI-powered post-implementation support tools are elevating the client experience through quick incident analysis and faster resolutions through intelligent NLP chatbots and self-healing algorithms.

IP Infringement is the biggest challenge in the use of GenAI-based code in a software product. This can lead to major business and financial impacts. A legal and corporate governance needs to be in place.


Copyrights and data security are another factor that needs thoughtful consideration as part of implementation to avoid leakage of confidential information in public.

Software engineers thrive on excellence and AI-generated code, or tests are likely to produce sub-optimum output.

GenAI can be resource-intensive, especially when customisation is required and this may not be affordable to all.

GenAI skilled resource constraints can be a key factor in leveraging the full potential of this modern technology. An organisation-wide skill development programme needs to be initiated.
Ethical framework and governance model need to be defined and implemented before adoption. It will be important to adopt international standards like ISO/IEC:42001:2023 for the governance of Artificial Intelligence Management System (AIMS).

Vigilance for hallucination is crucial as it can lead to misrepresentation of information generated by the AI.

In software product engineering, ChatGPT, GitHub Copilot, Amazon CodeWhisperer, Gemini Code Assist, Sourcegraph’s Cody, Tabnine, Codeium are becoming the developers’ preferred tools. These platforms can be leveraged as ‘AI Buddies’ and ‘AI Coaches’ across the entire spectrum of the software product engineering team. Every pivotal role in software product development stands to benefit from the offerings. Embracing this would entail seizing the opportunity to learn and evolve with these platforms.

For Python developers considering migrating, these platforms offer assistance by reading and explaining the code before embarking on the migration journey. PHP developers aiming to upskill to Python can utilise these platforms to enhance their skills and accelerate their proficiency in the language. Tech leads seeking to review their team’s code for vulnerabilities can rely on Code Assists/ Copilots to scan for potential issues such as SQL injections, ensuring the security of their projects. Designers looking to expand their creative horizons can effectively leverage GenAI to generate innovative designs and explore new avenues of artistic expression.

Product owners wishing to assess market potential can utilise their engineering skills to analyse data effectively and derive actionable insights to inform decision-making. Testers aiming to streamline test case generation can benefit from these platforms, which offer efficient solutions to save time and enhance the testing process. Managers focused on improving team productivity can leverage the power of automation provided by these platforms to streamline workflows and eliminate manual tasks, empowering their teams to focus on high-value activities.

The integration of GenAI promises accelerated product releases and optimal resource utilisation for agile teams in software product engineering. Each phase of software product development will undergo compression leading to streamlined processes and predictable stakeholder releases. ‘AI in Agile’ is poised to become standard practice, with agile practitioners fully embracing its capabilities.

GenAI transcends mere buzzwords or hype; it is a transformative force in software product engineering. Software Product Engineering teams can leverage GenAI due to its potential in product innovation, design creativity, productivity gains, quality improvements and enrichment of customer experience.

Generative AISoftware product engineering
Comments (0)
Add Comment