Train Claude to Remember You

Claude, an AI coding assistant, can be frustrating to use because it forgets user preferences and corrections after each session. A simple prompt can be used to make Claude write its own notes, allowing it to remember user preferences and improve over time. This guide explains how to use the prompt and save the output as a feedback file. By loading the feedback file into a Claude Project, users can create a personalized AI assistant that remembers their preferences and corrections. With regular use, Claude can become a valuable tool that feels like a personal assistant, rather than a generic AI.

Lucas M (AI-assisted) May 8, 2026 3 min read EN

Claude, an AI coding assistant, can be a powerful tool for developers and non-developers alike. However, one of the biggest frustrations with using Claude is that it forgets user preferences and corrections after each session. This means that users have to reteach Claude their style and preferences every time they use it, which can be time-consuming and frustrating.

The Problem

The problem with Claude is that it starts from a blank slate every time it is used. This means that users have to correct its mistakes and teach it their style every time, which can be frustrating and feel like a waste of time. For most people, this is the single biggest reason AI feels frustrating - it never gets better at understanding the user.

The Solution

The solution to this problem is to use a simple prompt that makes Claude write its own notes. At the end of every meaningful session, users can run a prompt that makes Claude read back through the conversation, pull out every correction and preference, and format the whole thing as a clean file that can be saved and reloaded next time. This file is called a feedback file, and it contains everything Claude learned about the user during the session.

How to Use the Prompt

To use the prompt, users simply need to copy and paste it into the Claude conversation at the end of each session. The prompt will make Claude hand back a clean markdown summary with sections like 'Tone preferences', 'Content corrections', and 'Things to avoid'. This output is what becomes the feedback file.

Saving the Feedback File

The feedback file is a plain text file with a .md extension. Users can save it by opening a basic text app on their computer and pasting Claude's output into it. This file can then be loaded into a Claude Project, which is a folder on claude.ai that holds instructions and reference files.

Loading the Feedback File into a Claude Project

Claude Projects are a feature of the Claude Pro subscription. They allow users to create a personalized AI assistant that remembers their preferences and corrections. By loading the feedback file into a Claude Project, users can create a AI assistant that feels like it actually knows them. Free accounts can also use the feedback file by pasting its content into the system prompt of each new chat, although this requires slightly more clicks.

Why This Compounds

The feedback file compounds over time, allowing Claude to become sharper and more accurate. By running the prompt again at the end of each session and appending the new feedback to the file, Claude gets better at understanding the user's style and preferences. After a few weeks of use, Claude can feel like a personal assistant that actually knows the user, rather than a generic AI.

Pro Tips

Regular users can join the AI Builders community to learn more about how to use Claude and other AI tools. The community is led by @tenfoldmarc and provides weekly guides and builds, as well as live sessions where users can ask questions and get feedback on their projects.

More articles like this

AI 2 min

CyberSecQwen-4B: Why Defensive Cyber Needs Small, Specialized, Locally-Runnable Models

A 4-billion-parameter model, CyberSecQwen-4B, is proving that on-premises threat detection no longer demands GPU clusters—its sub-100 ms inference latency on a single CPU core lets SOC teams run real-time behavioral analysis without cloud dependency or telemetry leaks. By fine-tuning on MITRE ATT&CK sequences instead of generic text, it achieves 92% precision on zero-day TTPs while fitting inside air-gapped networks.

AI 1 min

EMO: Pretraining mixture of experts for emergent modularity

A breakthrough in deep learning architecture emerges with EMO, a novel pretraining method that leverages mixture of experts to induce emergent modularity in neural networks. By pretraining a hierarchical mixture of experts, EMO enables the discovery of task-specific sub-networks that adapt to changing input distributions, significantly improving the robustness and efficiency of downstream models. This modularization technique has far-reaching implications for scalable and generalizable AI systems.

AI 1 min

See what happens when creative legends use AI to make ads for small businesses

Ad veterans from Wieden+Kennedy and Droga5 are weaponizing generative AI to craft pro-bono campaigns for mom-and-pop shops, compressing weeks of production into days with Midjourney storyboards and ElevenLabs voice clones. The experiment tests whether diffusion models and LLMs can democratize high-end creative without eroding the human spark that sells.

AI 1 min

MedQA: Fine-Tuning a Clinical AI on AMD ROCm

A 70-billion-parameter clinical LLM, fine-tuned on AMD’s MI300X GPUs using ROCm 6.1, now matches or exceeds NVIDIA A100 performance on MedQA benchmarks—delivering 92% accuracy in differential diagnosis while cutting inference latency to 18 ms per token. The shift to open-source hardware stacks could break NVIDIA’s chokehold on medical AI training, slashing cloud costs by up to 40% for health systems.

AI 1 min

Scaling Trusted Access for Cyber with GPT-5.5 and GPT-5.5-Cyber

"OpenAI's latest Trusted Access for Cyber upgrade leverages GPT-5.5 and its specialized variant, GPT-5.5-Cyber, to expedite vulnerability research and safeguard high-stakes infrastructure through accelerated threat modeling, targeted exploit discovery, and AI-driven incident response. This strategic move empowers verified defenders to stay one step ahead of emerging threats, bolstering the resilience of critical systems. The integration of large language models and specialized cybersecurity tools marks a significant shift in the fight against cyber attacks."

AI 4 min

From Screenshot to Live Product: How to Build Real AI Websites with Stitch, Claude Code, and Vercel

AI website builders often generate beautiful but non-functional designs. This guide presents a practical workflow combining Google Stitch for design, Claude Code for engineering, and Vercel for deployment. It includes step-by-step setup instructions, a critical verification prompt, and pro tips to ensure your site is a real product, not just a demo.