Memory & Personalization

Claude learns your preferences — here's how to make that work for you.

What is Claude's memory?

Claude's memory feature (Pro and above) allows Claude to remember information about you across conversations. Unlike the context within a single chat, memory persists between sessions — Claude can recall that you prefer Python over JavaScript, that you work as a data engineer at a fintech company, or that you like concise responses without preamble.

How memory is built

Claude builds memories automatically based on what you share in conversations. It looks for:

• Your role and profession ("I'm a data engineer at a bank")
• Your technical stack ("we use PostgreSQL and dbt")
• Your preferences ("I prefer concise answers", "always show me the code first")
• Your name when shared
• Your location for location-sensitive queries
• Your goals ("I'm preparing for a data engineering interview")

Viewing and managing your memories

1. Click your profile icon (bottom left)
2. Go to Settings → Memory
3. You'll see a list of what Claude currently remembers about you
4. You can delete individual memories or clear all
5. You can also add a memory manually by typing directly in the memory interface

Explicitly teaching Claude about you

The fastest way to personalize Claude is to tell it directly at the start of a conversation or in your Project Instructions:

About me:
- Name: Raman
- Role: Senior Data Engineer at a fintech startup
- Stack: Python 3.12, PostgreSQL 15, dbt-core, Apache Airflow 2.7, AWS
- Conventions: snake_case, 3NF in OLTP, star schema in DWH, BIGINT for all IDs
- Communication: concise answers, show code before explanation
- Currently learning: Spark, Delta Lake, Databricks

What Claude does NOT remember

• The full content of past conversations (only summaries are stored as memories)
• Code you shared in previous sessions (use Projects for persistent code context)
• Anything from Incognito mode chats
• Anything you explicitly delete from memory settings