Copy of National Archives Uses AI to Boost Access to Historical Records
Published Date::17/10/2024
AI agents are set to revolutionize software development, moving developers into review roles and enabling autonomous coding. However, challenges and questions remain.
Developers who are unimpressed by the early returns of generative AI for coding should take note the future of software development is heading toward an era where most code will be written by AI agents, with experienced developers primarily in review roles, according to Gartner.
Organizations and vendors are already rolling out AI coding agents that enable developers to automate or offload many tasks. More pilot programs and proofs-of-concept are expected to be launched in 2025, says Philip Walsh, senior principal analyst in Gartner’s software engineering practice.
By 2026, there will likely be more productive and mainstream levels of adoption of AI coding agents. People will have a better understanding of their strengths and weaknesses, and use cases where these agents can operate more autonomously. By 2027, this paradigm will take root, and engineers' workflows and skill sets will need to evolve and adapt.
In a recent press release, Gartner predicted that 80% of software engineers will have to reskill to fit into new roles created when generative AI takes over more programming functions.
These AI coding agents will be more advanced than the current AI coding assistants but will still need experienced programmers to check their work and tweak the code. The trend toward agentic AI, which emphasizes autonomous decision-making over simple content generation, will push the boundaries of current AI coding copilots, enabling AI-native software engineering to emerge.
While current AI coding assistants can write snippets of code, they often struggle to create software from scratch. However, the next generation of AI coding agents will be capable of handling higher-level goals and tasks, iteratively working through problems and finding solutions.
FAQS:
Q: What is AI-native software engineering?
A: AI-native software engineering refers to the practice of using AI agents to write the majority of code, with human developers primarily in review and fine-tuning roles.
Q: When will AI agents become mainstream in software development?
A: Gartner predicts that AI agents will start to be more widely adopted by 2026, with a significant shift in the industry by 2027.
Q: What are the main challenges of AI coding agents?
A: Main challenges include processing long contexts, identifying relevant code affected by changes, and maintaining the integrity of the codebase. Human developers are still needed to understand the broader context and fine-tune the code.
Q: How will the role of developers change with AI-native coding?
A: Developers will need to reskill and adapt to new roles, such as reviewing AI-generated code and handling complex software tasks that AI cannot yet manage.
Q: What are the potential impacts on the developer talent pipeline?
A: There are concerns that automating basic coding tasks could make it difficult to train junior developers and develop expertise in the field.