Skip to content

LLM apps problems

47 problems, rising first. The list grows every day.

ProblemThis week
Claude app gets 1 and 2 star reviewsClaude, ChatGPT, GitHub, Claude Code, Perplexity, PyTorch13d active+2Seen on App Store reviews, GitHub issues, Reddit, Community forums, Stack Overflow+2828 posts in the last 7 days, 2 the 7 days before (rising)
ChatGPT app gets 1 and 2 star reviewsChatGPT, Codex, OpenAI, Claude, Gemini, Perplexity11d activeSeen on App Store reviews, GitHub issues, Community forums+1717 posts in the last 7 days, 1 the 7 days before (rising)
Context windows run out on real workloadsOllama, vLLM, Cursor, Grok, ChatGPT, Codex8d active+1Seen on GitHub issues, Hacker News, Reddit, Community forums+66 posts in the last 7 days, 0 the 7 days before (rising)
Claude fails to return valid structured outputClaude, DeepSeek, OpenAI, Perplexity, Ollama, Hugging Face4d activeSeen on GitHub issues+33 posts in the last 7 days, 0 the 7 days before (rising)
Perplexity app gets 1 and 2 star reviewsPerplexity5d activeSeen on App Store reviews+99 posts in the last 7 days, 3 the 7 days before (rising)
RAG retrieval returns irrelevant or missing contextGitHub, Vercel, SQL Server, Anthropic, OpenAI, Perplexity6d active+1Seen on GitHub issues, Hacker News, DEV, Reddit+33 posts in the last 7 days, 1 the 7 days before (rising)
Anthropic users keep filing bugsAnthropic, OpenAI, Perplexity, DeepSeek, Ollama, Hugging Face7d activeSeen on GitHub issues, App Store reviews, Reddit+66 posts in the last 7 days, 0 the 7 days before (rising)
LLMs give confidently wrong answersGitHub, PyPI, Claude, Grok, ChatGPT, Power BI9d activeSeen on Hacker News, Reddit, Bluesky+44 posts in the last 7 days, 4 the 7 days before (steady)
OpenAI users keep asking for helpOpenAI, LlamaIndex, PyPI, vLLM, GitHub, MLflow9d active+1Seen on Community forums, GitHub issues, Stack Overflow, Reddit+55 posts in the last 7 days, 2 the 7 days before (rising)
Gemini gives confidently wrong answersGemini, GitHub, VS Code, Model Context Protocol, npm, YouTube3d activeSeen on GitHub issues, Hacker News+22 posts in the last 7 days, 0 the 7 days before (rising)
Grok users keep asking for helpGrok, Cursor, Claude6d activeSeen on Community forums, App Store reviews+44 posts in the last 7 days, 0 the 7 days before (rising)
Trouble with Anthropic raisesAnthropic, OpenAI, Perplexity, DeepSeek, Ollama, Hugging Face2d activeSeen on GitHub issues+22 posts in the last 7 days, 0 the 7 days before (rising)
Trouble with Gemini studioGemini3d activeSeen on Community forums+33 posts in the last 7 days, 0 the 7 days before (rising)
No reliable way to evaluate LLM app qualityllama.cpp, NVIDIA, Hugging Face, Qwen, GitHub, MLflow15d activeSeen on GitHub issues, Hacker News, Reddit+33 posts in the last 7 days, 3 the 7 days before (steady)
API rate limits and usage caps block real workDeepSeek, Ollama, LangChain, Model Context Protocol, OpenRouter, Anthropic12d active+1Seen on GitHub issues, Hacker News, Reddit, Community forums+33 posts in the last 7 days, 3 the 7 days before (steady)
Claude runs out of context on real workClaude, Codex, ChatGPT, LlamaIndex2d activeSeen on Hacker News, GitHub issuesUnverified+22 posts in the last 7 days, 0 the 7 days before (new)
Model Context Protocol users keep filing bugsModel Context Protocol, Microsoft Azure, Jira, Salesforce, ServiceNow, npm11d activeSeen on GitHub issues, Reddit+44 posts in the last 7 days, 1 the 7 days before (rising)
Qwen users keep asking for helpQwen, NVIDIA, Hugging Face, vLLM, PyTorch, GitHub3d activeSeen on Reddit, GitHub issues+33 posts in the last 7 days, 0 the 7 days before (rising)
Models fail to return valid structured outputGitHub, Hugging Face, Ollama, LlamaIndex, ChatGPT, Anthropic6d activeSeen on GitHub issues, Hacker News, Reddit+22 posts in the last 7 days, 2 the 7 days before (steady)
Gemini users keep filing bugsGemini, GitHub, VS Code16d activeSeen on GitHub issues, Community forums+44 posts in the last 7 days, 5 the 7 days before (steady)
LlamaIndex users keep filing bugsLlamaIndex, GitHub9d activeSeen on GitHub issues, Stack Overflow+22 posts in the last 7 days, 2 the 7 days before (steady)
LangChain users keep filing bugsLangChain, OpenAI, Perplexity, DeepSeek, Ollama, Anthropic3d activeSeen on GitHub issues+22 posts in the last 7 days, 2 the 7 days before (steady)
AI agents are unreliable, loop or stall midwayModel Context Protocol, Claude, LangChain, Anthropic, OpenAI, Perplexity8d active+1Seen on GitHub issues, Hacker News, Bluesky, Reddit+11 post in the last 7 days, 5 the 7 days before (fading)
Claude Code runs out of context on real workClaude Code, Claude, Codex, Oracle, Cursor, Windsurf4d activeSeen on Hacker News, GitHub issues+11 post in the last 7 days, 0 the 7 days before (steady)
Model quality silently gets worse between versionsCodex, Cursor, JetBrains, AWS, Microsoft Azure, Google Cloud4d activeSeen on Hacker News, Bluesky, Reddit+11 post in the last 7 days, 1 the 7 days before (steady)
DeepSeek users keep filing bugsDeepSeek, Hugging Face, Ollama, vLLM, GitHub, LangChain3d activeSeen on GitHub issues+11 post in the last 7 days, 0 the 7 days before (steady)
Anthropic agents loop, stall or fail midwayAnthropic, LangChain, Codex, Perplexity, DeepSeek, Ollama2d activeSeen on GitHub issuesUnverified+11 post in the last 7 days, 0 the 7 days before (steady)
Model Context Protocol gives confidently wrong answersModel Context Protocol, PyPI, Claude, Cursor, GitHub2d activeSeen on GitHub issuesUnverified+11 post in the last 7 days, 0 the 7 days before (steady)
MLflow rate limits and usage caps block workMLflow2d activeSeen on GitHub issuesUnverified+11 post in the last 7 days, 1 the 7 days before (steady)
Trouble with ChatGPT windowsChatGPT, Codex4d activeSeen on GitHub issues+00 posts in the last 7 days, 0 the 7 days before (fading)
Trouble with LangChain windowsLangChain, Anthropic, OpenAI, Perplexity, DeepSeek, Ollama2d activeSeen on GitHub issues+00 posts in the last 7 days, 1 the 7 days before (fading)
GitHub rate limits and usage caps block workGitHub, Vercel, npm3d activeSeen on GitHub issues, Hacker News+00 posts in the last 7 days, 1 the 7 days before (fading)
Trouble with DeepSeek flashDeepSeek, vLLM, Hugging Face, NVIDIA, GitHub, Ollama3d activeSeen on GitHub issues+00 posts in the last 7 days, 2 the 7 days before (fading)
Trouble with Claude desktopClaude, Claude Code, OpenAI, GitHub, npm, PyPI3d activeSeen on GitHub issues+00 posts in the last 7 days, 2 the 7 days before (fading)
Trouble with Gemini flashGemini3d activeSeen on Community forums, GitHub issues+00 posts in the last 7 days, 1 the 7 days before (fading)
OpenAI fails to return valid structured outputOpenAI, LlamaIndex, PyPI, Venice, Gemini, MLflow2d activeSeen on GitHub issues, Hacker News+00 posts in the last 7 days, 0 the 7 days before (fading)
ChatGPT rate limits and usage caps block workChatGPT, Codex, Model Context Protocol3d activeSeen on GitHub issues+00 posts in the last 7 days, 0 the 7 days before (fading)
Codex rate limits and usage caps block workCodex, ChatGPT3d activeSeen on GitHub issues+00 posts in the last 7 days, 0 the 7 days before (fading)
Trouble with OpenAI cloudOpenAI, Ollama, DeepSeek, Z.ai3d activeSeen on GitHub issues+00 posts in the last 7 days, 0 the 7 days before (fading)
Trouble with LangChain coreLangChain, GitHub2d activeSeen on GitHub issues+00 posts in the last 7 days, 0 the 7 days before (fading)
GitHub Copilot gives confidently wrong answersGitHub Copilot, VS Code2d activeSeen on GitHub issues, RedditUnverified+00 posts in the last 7 days, 2 the 7 days before (fading)
LLM responses are too slow for usersCapacity, Claude, Claude Code, Llama, Ollama2d activeSeen on DEV, GitHub issuesUnverified+00 posts in the last 7 days, 1 the 7 days before (fading)
Cloudflare rate limits and usage caps block workCloudflare, Supabase, n8n, Deutsche Telekom2d activeSeen on GitHub issuesUnverified+00 posts in the last 7 days, 2 the 7 days before (fading)
Kubernetes agents loop, stall or fail midwayKubernetes, Dagster, GitHub, AWS, Docker2d activeSeen on DEV, GitHub issuesUnverified+00 posts in the last 7 days, 1 the 7 days before (fading)
RAG on LangChain returns irrelevant or missing contextLangChain, Airbyte, MLflow1d activeSeen on GitHub issuesUnverified+00 posts in the last 7 days, 0 the 7 days before (fading)
RAG on LlamaIndex returns irrelevant or missing contextLlamaIndex, Cloudflare, Model Context Protocol, GitHub, PyPI2d activeSeen on GitHub issuesUnverified+00 posts in the last 7 days, 0 the 7 days before (fading)
Claude Code rate limits and usage caps block workClaude Code, Claude2d activeSeen on GitHub issuesUnverified+00 posts in the last 7 days, 0 the 7 days before (fading)

Rising problems by email

Mondays: the problems in data, tech and AI that grew fastest that week.

Double opt-in. Unsubscribe any time.