More than 9 in 10 developers are excited about AI’s impact on their careers, and an overwhelming 96% expect it to change the developer experience for the better. Salesforce’s latest State of IT survey of software development leaders also reveals that 86% believe AI agents will become as essential to app development as traditional software tools.
The large global study of more than 2,000 software development leaders, including 50 from Ireland, highlights near unanimous excitement about agentic AI. Developers are not only looking to agents to unlock greater efficiency and productivity, but 92% believe agentic AI will help them advance in their careers. Some developers, however, believe that they, as well as their organisations, need more training and resources to build and deploy a digital workforce of AI agents.
Why it matters: Developers have often been painted as wary of AI, but new research reveals developers are enthusiastic about the industry’s shift to AI agents. The arrival of agentic AI provides developers with the opportunity to focus less on tasks like writing code and debugging and grow into more strategic, high-impact work. And with developers increasingly using agents powered by low-code/no-code tools, development is becoming faster, easier, and more efficient than ever — regardless of developers’ coding abilities.
Salesforce perspective: “AI agents are revolutionising the way developers work, making software development faster, more efficient, and more enjoyable. This powerful digital workforce streamlines development by assisting with writing, reviewing, and optimising code — unlocking new levels of productivity. By automating tedious tasks like data cleaning, integration, and basic testing, AI agents free developers to shift their focus from manual coding to high-value problem-solving, architecture, and strategic decision-making.” – Alice Steinglass, EVP & GM, Platform, Integration and Automation
By the numbers:
Developers are enthusiastic about agentic AI and its impact on their career
Respondents are excited for agents to take on simple, repetitive tasks, freeing them up to focus on high-impact projects that contribute to larger business goals.
- 96% of developers are enthusiastic about AI agents’ impact on the developer experience.
- Developers are most eager to use AI agents for debugging and error resolution, then for generating test cases and building repetitive code.
- The arrival of agentic AI comes at a time where 92% of developers are looking to measure their productivity based on impact over output.
- With the help of AI agents, developers believe they’ll focus more on high-impact projects like AI oversight and architecting complex systems.
Low-code/no-code tools help developers unlock greater productivity, regardless of coding skills
With agents powered by low-code/no-code tools, developers of all levels can now build and deploy agents. Respondents believe these tools will help democratise and scale AI development for the better.
80% of developers using agentic AI currently use low-code/no-code tools.
78% of developers say that low-code/no-code tools can help democratise AI development.
90% of developers say that the use of low-code/no-code app development tools can help scale AI development.
The customer view: Secret Escapes offers discounted luxury travel hotels and luxury across the globe and is using Agentforce to deliver more efficient customer support to unique and last-minute requests from its travellers. With Salesforce’s low-code Agent Builder tools, the Secret Escapes developer team was able to build, test, and launch their AI agent in just two weeks, compared to the six months it had previously taken the company to build and train a bot.
Developers are eager for more resources to build AI agents
Developers say updated infrastructure and more testing capabilities and skilling opportunities are critical as they transition to building and deploying AI agents.
- Infrastructure Needs: Many developers (84%) believe their organisation needs to update their infrastructure to build/deploy AI agents.
- 42% of developers say their data quality and accuracy isn’t sufficient for the successful development and implementation of agentic AI.
- Testing Capabilities: Nearly half (46%) of developers say their testing processes aren’t fully prepared to build and deploy AI agents.
- Skills and Knowledge: Almost 7 in 10 (68%) developers believe AI knowledge will soon be a baseline skill for their profession, but over half (54%) don’t feel their skillsets are fully prepared for the agentic era.
- Survey respondents identified training on technical AI skills and redefining current roles as the most important areas for employers to provide support.
Go deeper:
- Explore CIO insights on AI implementation in Salesforce’s CIO Dilemma research
- Discover the latest in AI adoption from Slack’s Workforce Index
- Learn how developers can use low-code tools with Salesforce Platform to build agents
- Register for TDX March 5-6, 2025
Data is sourced from a software development leader segment of a double-anonymous survey of IT decision-makers conducted from December 24, 2024 through February 3, 2025. Respondents represent Australia, Belgium, Brazil. Canada, Denmark, Finland, France, Germany, India, Indonesia, Ireland, Israel, Italy, Japan, Mexico, the Netherlands, New Zealand, Norway, Portugal, Singapore, South Korea, Spain, Sweden, Switzerland, Thailand, the United Arab Emirates, the United Kingdom, and the United States.
Additional data was collected among nearly 250 frontline developers, in partnership with YouGov, in the United States from February 14 through February 21, 2025.
