The software engineers are not alright.
AI was supposed to make programming easier. Siddhant Khare said that while AI tools have made him more productive, his job is now harder than ever.
“We used to call it an engineer, now it is like a reviewer,” Khare told Business Insider. “Every time it feels like you are a judge at an assembly line and that assembly line is never-ending, you just keep stamping those PRs.”
Khare wrote a lengthy essay titled “AI fatigue is real and nobody talks about it.” In it, he wrote that AI fatigue is “the kind of exhaustion that no amount of tooling or workflow optimization could fix.”
“I shipped more code last quarter than any quarter in my career,” he wrote. “I also felt more drained than any quarter in my career.”
Khare, who is a software engineer at ONA, makes AI tools himself. He said he’s not against AI, he just wants people, including software engineers like himself, to find a more sustainable way to use the technology.
The issue, Khare wrote, is that AI creates a paradox of productivity that falls almost entirely on the user to solve by reducing the cost of production but increasing “the cost of coordination, review, and decision-making.”
His typical day is no longer devoted to deep focus on a single design problem, Khare wrote, but rather to context switching between as many as six different problems.
“Each one ‘only takes an hour with AI.’ But context-switching between six problems is brutally expensive for the human brain,” he wrote. “The AI doesn’t get tired between problems. I do.”
Early reactions show Khare isn’t alone. Fellow engineers have shared their thoughts on X, Hacker News, and Lobsters. Their experiences echo what Khare wrote: vibe coding can lead to burnout vibes.
“I joke that I’m on the ‘Claude Code workout plan’ now,” one user wrote on Hacker News. “Standing desk, while it’s working I do a couple squats or pushups or just wander around the house to stretch my legs. Much more enjoyable than sitting at my desk, hands on keyboard, all day long. And taking my eyes off the screen also makes it easier to think about the next thing.”
Others make their day sound like the “C’mon do something meme” as they wait for a coding agent to respond with results, which they will then have to edit and refine.
“I used to lose myself in focused work for hours,” another user wrote. “That’s changed. Now I’m constantly pulled away, and I’ve noticed the pattern: I send a prompt, wait for the reply, and drift into browsing.”
On Monday, Harvard Business Review published a report on an eight-week study at a US tech company with 200 employees that “discovered that AI tools didn’t reduce work, they consistently intensified it.”
“Once the excitement of experimenting fades, workers can find that their workload has quietly grown and feel stretched from juggling everything that’s suddenly on their plate,” the report continued. “That workload creep can in turn lead to cognitive fatigue, burnout, and weakened decision-making.”
AI tool FOMO and skill atrophy
Khare said another struggle he has had is keeping up with the number of updates from OpenAI, Anthropic, and other AI companies as they push their popular models and agents.
“Once you open your laptop, your phone either you go to Slack, Twitter, LinkedIn, GitHub, and these are all things that are like ‘Look at me,'” Khare said.
The companies need to keep innovating to stay apace in the AI race, but for programmers like himself, Khare said, the updates create a sense of panic about missing out on what could be the next greatest thing.
“I was spending weekends evaluating new tools,” he wrote. “Reading every changelog. Watching every demo. Trying to stay at the frontier because I was terrified of falling behind.”
Ultimately, Khare wrote that the scariest thing was how AI was changing him. He said he struggled when someone asked him “to reason through a concurrency problem on the whiteboard” with no laptop or AI.
“It’s like GPS and navigation. Before GPS, you built mental maps. You knew your city. You could reason about routes,” he wrote. “After years of GPS, you can’t navigate without it. The skill atrophied because you stopped using it.”
Andrej Karpathy, Tesla’s former AI head and the engineer who coined the term “vibe coding” last year, recently wrote that while the industry was seeing a “phase shift” among increasingly impressive AI coding tools, he’d also begun to notice skill atrophy.
“I’ve already noticed that I am slowly starting to atrophy my ability to write code manually,” Karpathy wrote.
Finding the right answers to AI usage
Determined to help himself, Khare wrote that he had come up with a list of ground rules to rein in his AI use.
He’s still struggling to hold himself to it. He said almost completely ignoring the AI conversation over his 14-day holiday break was a start. Khare said that while he wrote that having a 30-minute timer on AI use is helpful, it wasn’t intended to be a strict limit.
The engineer also said that AI companies themselves could help as well. He compared the feeling of being just one prompt away from a perfect answer to the way addictive games try to keep users playing.
“You need to keep some sort of guardrails for the humans, so they don’t self-distruct themselves,” he said.
Read the full article here
