What AI models does the agent use?

The agent has access to Claude, DeepSeek, Gemini, and other top models. It automatically routes to the best model for each task — you never have to think about API keys or token limits.

Yes. Each agent runs in an isolated environment. We don't train on your data, and you can export or delete everything at any time. The underlying engine (GoGogot) is open-source, so you can audit exactly how it works.

What can the agent actually do?

Browse the web, process files (CSV, PDF, images), run scheduled tasks, remember context across sessions, and communicate via Telegram or Slack. Think of it as a capable junior employee, not a chatbot.

How does memory work?

The agent maintains persistent memory across sessions. It remembers your preferences, past conversations, and task context. You can also explicitly tell it to remember or forget things.

Can I self-host the agent?

Absolutely. GoGogot is 100% open-source. Run it on your own hardware for free. cowork.ink is the managed version — we handle servers, model costs, updates, and uptime so you don't have to.

Is there a free trial?

We offer a 7-day money-back guarantee. Spin up an agent, give it real tasks, and if it doesn't save you time, we'll refund you — no questions asked.

How does an AI agent remember my preferences across sessions?

Most AI agents use a persistent memory layer — a structured external store where the agent writes observations after each session and retrieves them at the start of the next. This is different from fine-tuning; no model weights change. Your preferences live in a database, not the model itself. See our deep dive on [AI agent memory](/blog/ai-agent-memory/) for the full breakdown.

What is the difference between a self-learning AI agent and a regular chatbot?

A regular chatbot starts fresh every conversation with no knowledge of your history. A self-learning agent carries episodic and semantic memories across sessions, adjusts its behavior based on your feedback, and over time surfaces what you need without being prompted. The gap widens every week you use it.

Do I need to manually retrain an AI agent every time my preferences change?

No. Modern agents that use RAG or persistent memory don't require retraining. They update their memory store automatically when your behavior shifts — the next session simply retrieves the newer memories. Fine-tuning (which does require retraining) is rarely used for personal preference learning because it's expensive and slow.

How long does it take for an AI agent to noticeably improve from my habits?

Agents using RAG or persistent memory can improve within a single session — the agent logs your preferences and applies them immediately in subsequent turns. Behavioral patterns (your communication style, priorities, recurring tasks) typically become reliably encoded after 5–10 sessions of normal use.

Can I control what an AI agent learns about me?

Yes, in well-designed systems. Good agent platforms let you view, edit, and delete stored memories. This is an important privacy control — if the agent has learned a wrong habit or outdated preference, you should be able to correct it directly. Always check whether your agent platform gives you memory transparency and control.

How AI Agents Learn Your Habits and Get Better Every Day

Discover how AI agents learn user habits through memory, RAG, and feedback loops. EXPLAINED simply — see how agents get smarter every session. Try cowork.ink free.

Frequently Asked Questions

How does an AI agent remember my preferences across sessions?: Most AI agents use a persistent memory layer — a structured external store where the agent writes observations after each session and retrieves them at the start of the next. This is different from fine-tuning; no model weights change. Your preferences live in a database, not the model itself. See our deep dive on [AI agent memory](/blog/ai-agent-memory/) for the full breakdown.
What is the difference between a self-learning AI agent and a regular chatbot?: A regular chatbot starts fresh every conversation with no knowledge of your history. A self-learning agent carries episodic and semantic memories across sessions, adjusts its behavior based on your feedback, and over time surfaces what you need without being prompted. The gap widens every week you use it.
Do I need to manually retrain an AI agent every time my preferences change?: No. Modern agents that use RAG or persistent memory don't require retraining. They update their memory store automatically when your behavior shifts — the next session simply retrieves the newer memories. Fine-tuning (which does require retraining) is rarely used for personal preference learning because it's expensive and slow.
How long does it take for an AI agent to noticeably improve from my habits?: Agents using RAG or persistent memory can improve within a single session — the agent logs your preferences and applies them immediately in subsequent turns. Behavioral patterns (your communication style, priorities, recurring tasks) typically become reliably encoded after 5–10 sessions of normal use.
Can I control what an AI agent learns about me?: Yes, in well-designed systems. Good agent platforms let you view, edit, and delete stored memories. This is an important privacy control — if the agent has learned a wrong habit or outdated preference, you should be able to correct it directly. Always check whether your agent platform gives you memory transparency and control.