The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
RAG is not the bottleneck. Your data is.
Everyone's tweaking embeddings, swapping models, A/B testing prompt formats. Then they ship. Then the bot still hallucinates.
Here's where I usually see things break:
🚫 What people debug
• "Maybe gpt-4o is the wrong model"
• "Try a different embedding model"
• "Increase top_k to 20"
• "Add a reranker, that'll fix it"
✅ What actually fixes it
• Chunk by meaning, not by 500 tokens
• Keep the source structure (headings, tables, lists)
• Write summaries of long sections, embed those too
• Drop duplicates, drop legal footers, drop nav menus
• Test on real user questions, not your own
Most "AI accuracy" problems are data problems pretending to be model problems.
Build the pipeline properly once, and the same model gives you 2x better answers.
If you're stuck on this, drop a comment with your case. Happy to debug for free, I learn from each one.
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Related posts
Introducing AgroGPT 🌱 An AI agriculture assistant built for farmers across Bangladesh.
AgroGPT brings farming guidance and local information together in one mobile experience:
🌾 Six AI specialists covering crops, cattle, poultry, aquaculture, soil, and research
🌦️ Weather forecasts focused on crop risks
📊 Live bazar prices to support market decisions
📰 Agricultural news and updates
💬 AI chat for everyday farming questions
The design pairs a fresh green palette with clear navigation and dedicated specialist tools, making a wide range of information easier to explore.
What feature would you find most useful in an agriculture app?
Of course, AI now can do more than basic understanding of rough sketches or shapes, it is far beyond what the public frontier AI models we are able to access can even do, we are far too limited from the use if its full power, no complaints from me though, we should strive to support AI safety-measures to avoid catastrophe.
What do you guys use AI for?
Created using AI with a strong focus on accurately reproducing the original reference while maintaining realistic details, lighting, textures, and visual consistency.
