LLMs May Have Helped My RSI

TL;DR

Prime Big Deal Days · Oct 6–7Offer from Amazon

Get everyday helpers delivered free — and shop member deals

• Fast, free delivery on millions of items
• Access to Prime Big Deal Days deals on October 6–7
• Prime Video, Amazon Music and more included

Start your free Prime trial →Free trial for eligible customers · Cancel anytime

As an affiliate, we earn on qualifying purchases.Hide

A software developer says his repetitive strain symptoms have become less severe as AI coding agents take on some code-writing tasks. His account is personal, and it does not establish that LLM use caused the change; seniority, work habits and other factors may also explain it.

Vaughan Hilts reported that serious, burning arm pain has become uncommon as AI coding agents take on some of his code-writing work. In a blog post published October 5, 2026, the developer said the change may be related to spending less time manually coding, but he did not claim that LLMs caused the improvement.

Hilts wrote that he had dealt with what he and his doctors understood as repetitive strain injury (RSI) since around 2017. He described periods of burning forearm pain after heavy keyboard use, including long stretches of coding. He said symptoms had varied over the years and that exercise, breaks, ergonomic keyboards, reduced keyboard time and changes to his typing technique had helped him manage them.

He said the work he does has changed gradually during 2026. AI agents now handle some mechanical refactors and feature implementation that he might previously have typed and debugged himself over several hours. He estimates that a feature that could take four to five hours of manual work may now require prompts followed by about 30 minutes of manual refinement. He said he still writes code when needed, while spending more time on design specifications and reviewing code.

Hilts reported that he has shipped more code this year than in any previous year, while spending more of his keyboard time writing prose and less on code entry. He also noted that his increased seniority may mean more meetings and less hands-on typing. The post is an account of his own experience, not a clinical assessment or controlled comparison of symptoms before and after using AI tools.

AI Changes the Work Behind Coding

Hilts’s account points to a possible workplace effect of coding assistants beyond speed: they may change the kinds of physical tasks developers perform. If agents take on repetitive code entry, some workers could spend less time reaching for symbols and modifier keys and more time writing specifications or reviewing output. That possibility may matter to people whose work involves long hours at a keyboard.

But one person’s experience cannot show that LLMs prevent or relieve RSI. Hilts described several changes that could affect symptoms, including established habits such as taking breaks and exercising, as well as changes in his role and workload. His post supports a limited conclusion: he noticed less serious pain during a period when AI tools were increasingly part of his workflow, and he is unsure why.

Years of Keyboard-Related Pain

Hilts said he had tried a range of approaches over several years, including ergonomic keyboards, breaks, less time at the keyboard, exercise and retraining his typing technique. He found that some work was more physically demanding than other work. Writing design documents could involve long periods of typing with little discomfort, he said, while coding sessions often required frequent use of punctuation, function keys, navigation keys and modifiers.

He also described having relatively small hands and difficulty reaching certain key combinations, including combinations he encountered in the video game Final Fantasy XIV. These details help explain why he suspects the type of keyboard movement may have mattered to him, but they do not establish a medical cause for his symptoms or their improvement.

Cause of Symptom Change Is Unknown

The post does not provide a symptom diary, medical records, a diagnosis history or a comparison of keyboard time before and after adopting coding agents. It is not clear how much Hilts’s total keyboard use changed, how often he uses the tools, or whether his symptoms were assessed by a clinician during this period. He also mentions other possible influences, including exercise, breaks, typing habits, seniority and time spent in meetings.

For those reasons, his report cannot establish that LLMs caused his pain to improve, or that AI coding tools would have the same effect for other people. It also does not show that the tools treat or prevent RSI. The account is personal testimony, and the relationship between changing his workflow and changing symptoms remains uncertain.

No Follow-Up Study Reported

Hilts’s post does not announce a study, clinical follow-up or a plan to track his symptoms against keyboard use. Readers should treat the account as an individual observation rather than evidence of a general health benefit from AI-assisted coding. Whether his reduced pain continues, and what role—if any—the coding agents play, is not established in the report.

Anyone experiencing persistent or worsening pain should seek advice from a qualified health professional rather than relying on changes in software or work routines as treatment. Further evidence would be needed to determine whether AI tools meaningfully change physical strain for developers across different jobs and working patterns.

Key Questions

What did Vaughan Hilts report?

Hilts said he had experienced less serious arm pain in recent months after years of symptoms he associated with extensive keyboard use. He noted that AI agents now handle some coding tasks he once completed manually.

Does the post prove LLMs relieved his RSI?

No. It describes one person’s experience and does not establish cause and effect. Hilts himself raised other possible explanations, including changes in seniority, meetings, exercise and typing habits.

How have AI coding agents changed his work?

Hilts said he uses agents for some mechanical refactors and feature work, then prompts and manually refines the results. He said this can replace several hours of typing and debugging with a shorter period of review and refinement.

Does the report say AI coding tools treat RSI?

No. The post is not medical research and makes no general treatment claim. It describes a possible link in Hilts’s own work experience; people with persistent pain should consult a qualified health professional.

Source: hn