It’s been a rough stretch for the software engineer who listens to podcasts.
This month, Meta CEO Mark Zuckerberg said on The Joe Rogan Experience that the company will probably, in 2025, “have an AI that can effectively be a sort of mid-level engineer that you have at your company that can write code.”
Salesforce CEO Marc Benioff said in a December episode of the The Twenty Minute VC podcast that the company’s business plan, at least at that time, involved “not adding anymore software engineers,” due to productivity gains from AI tools.
If that’s not enough for an engineer to hit the “go back 30 seconds” button forever, a recent study from software-delivery platform Harness claims that 90% of a surveyed 500 engineering leaders and developers “are concerned that AI tools will replace developers.”
Analysts of the software market that spoke to IT Brew, however, aimed to put developers’ minds at ease. AI assists with tasks like automatic code completion, but the technology sometimes misses on tasks like bug reports; three pros who spoke to us told engineers to stay calm as execs catch up on AI’s strengths and weaknesses.
“[Executives] still, I think, have a higher expectation of what AI will deliver than what developers have found that it does deliver,” Andrew Cornwall, senior analyst at Forrester, told us.
Research and advisory firm Forrester predicted in October 2024 that “at least one large organization will try to replace 50% of its developers with AI and fail.”
Today’s AI tools assist with tasks like autocompleting code, test generation, or adding a virtual pair programmer to answer questions or perform commands.
For Dave Micko, senior director at market-intelligence firm Gartner, AI can set a developer up on the right path, but the tech lacks a certain quality found in most software engineers he’s known: the ability to ask uncomfortable questions.
Top insights for IT pros
From cybersecurity and big data to cloud computing, IT Brew covers the latest trends shaping business tech in our 4x weekly newsletter, virtual events with industry experts, and digital guides.
Let’s say a front-end developer uses a component library that has a drop-down menu of all 50 states. AI tools may use that kind of component without understanding the business context. What if the US company is planning to expand south?
“AI doesn’t know that we're expanding into Mexico, and I do,” Micko told IT Brew.
“What I need to do is take my human input, my ability to see patterns and they out, see outside and understand the strategic imperatives of the organization that I'm working for, and apply those types of value into the code that I’m creating,” he said.
Seth Robinson, VP of industry research at CompTIA, sees an essential role for the software developer as AI makes software development easier, as companies demand more software, and as expertise is required to connect those assets.
Many departments have their own niche software-as-a-service tools, he said, that need to be tied together as teams in the organization work together and require workflow capabilities.
CompTIA’s 2025 Tech Jobs report revealed “software developers and engineers” leading all tech positions in job post activity. December saw 40,455 now-hiring instances.
IT Brew has reported on one firm’s optimism in a software-engineering rebound.
Jayesh Govindarajan, EVP of Salesforce AI, said engineers will need to reimagine their roles and focus on emerging AI-focused skills like labeling data, crafting inputs for an agentic system, and reviewing AI outputs.
What assurance can Govindarajan provide a worried software engineer?
“I think the assurance is in changing with the times and building new skill sets,” he told us.
Salesforce currently displays 740 matching jobs for “software engineer.”