Products · Broad category
LLM coding assistants
Raised as a usability problem by 13 people on Hacker News over 3 years (2024–2026), in 11 threads.
This is a broad category rather than one product, so its issues aren't ranked on the home page.
People raised many different problems with LLM coding assistants, but no single problem has yet been raised by 3 people in 3 different months and threads.
Unreviewed issues were grouped by machine: Claude Haiku labelled each comment as a usability complaint, named what it's about and summed up the problem in a line, and a small model grouped similar problems. Nobody has checked them by hand. Expect mistakes.
The newest 13 complaints
- Agents promote excessive delegation, which makes it hard to enter a flow state2026-07-04
- Slow responses force splitting tasks into tiny prompts, hindering multitasking across agents2026-07-03
- Same prompt yields different APIs each time, forcing repeated prompt tweaking.2026-02-25
- Mixes methods from different library versions, producing unusable code2026-01-24
- Gives bogus advice, loops back to the same wrong answer after corrections2025-12-11
- Breaks in unintuitive ways when given instructions, feels borked from a UX perspective2025-10-09
- Breaks in unintuitive ways when told to do advertised tasks; footguns left in2025-10-09
- Breaks in unintuitive ways, requiring arcane knowledge to use properly2025-10-09
- Infuriating to use unless it one-shots simple coding tasks2025-09-12
- Asking for whole files takes long, wastes tokens, and breaks or truncates unrelated code2025-08-10
- Requires many iterations, waiting, and reading generated code, and sessions suddenly derail2025-03-31
- Ignores the user's stated preferences and style, and debugs by adding prints instead of fixing bugs2025-03-01
- Code generators have usability so steep that writing code yourself is easier2024-05-13