Claude와의 첫 대화

Imagine this: you're looking at a blank page, wrestling with a big, complex project, and not sure where to start. What if you had a partner who could help you break it down? Claude is a powerful, intelligent collaborator that amplifies your capabilities across all of your work. Claude brings AI intelligence, but you bring the context and expertise that makes the work meaningful.

Let's explore chatting with Claude, beginning with the sidebar. Here you will encounter icons for creating new chats and jumping back into old ones. There are options for projects where you can organize your conversations with persistent context and custom instructions, and then artifacts which allow you to turn ideas into shareable apps, tools, or content. No matter what you hope to accomplish, Claude offers support to achieve your goals.

All interactions with Claude begin with a prompt, and these prompts, combined with other context, impact its response. The best approach when speaking to Claude is like you would a coworker: naturally, concisely, and conversational. But you may ask, what is a good prompt?

Before your next conversation with Claude, consider a few things. One, setting the stage. What is your role and what are your objectives? Is there context about your work that Claude should know about? Two, defining the task. What action do you want Claude to take? Do you want Claude to write, analyze, build, etc.? Three, specifying rules. Is there a style or tone you want Claude to use? Are there examples that you can attach to show Claude?

Let's review this prompt below. First, we're setting the stage by telling Claude this is for an investor pitch deck for a new indie streaming app. That's the context and the objective. Then, we define the task clearly by providing relevant details. Finally, we specify the rules by using current web research with citations, structuring it as a professional report, which tells Claude exactly what style and format we need. Perfect.

With your basic prompt in place, now comes the real power move: adding your specific context to make Claude's response truly tailored to your needs. When you upload relevant documents or background information into a specific chat, Claude considers that context in its response, whether that's company documents, project files, or background research. Think of it as a shortcut so that Claude can understand what your needs are. Claude can intake and output many file types such as PDF, CSV, docs, and more.

In the search and tools menu, you can level up your experience yet again by giving Claude access to more tools, to pull in more context, and it will pull in these tools as needed while solving your request. You might toggle on options like web search to let Claude research current market data, or use connected data sources to pull from Google Drive, allowing access to real-time information. Claude will choose the right approach.

While those options directly impact the information Claude has access to, there are a few more ways to customize how Claude responds. You could choose the best Claude model for your task. To use the model selector, start chatting with Claude or open an existing chat. The selected model will appear under your text input. To change it, click on the model name and choose which model of Claude you'd like to chat with instead. Changing the model will result in a new chat.

Claude currently supports numerous models with distinct capabilities intended to meet specific user's needs. Use the Claude Opus model for your most complex tasks. It's the largest scale model with hybrid reasoning capabilities, for example, running a multi-step financial analysis. Claude Sonnet is the best choice for everyday users and is our recommended default model. It balances sophisticated capabilities with cost effectiveness. Think of it like choosing the right tool for the job. You can always switch models based on what you're trying to accomplish.

Extended thinking mode is particularly valuable for complex thinking and analysis, though it increases latency and may not be necessary for simple questions and everyday conversations. To get the fastest output for more straightforward tasks, you can start with Sonnet models and leave thinking mode off. If you need more support, you can change the model or turn on thinking mode.

Claude Sonnet enables Claude to conduct systematic multi-angle investigations by automatically breaking down complex questions and exploring hundreds of sources to deliver comprehensive, cited reports in five to 45 minutes. Claude does the research for you while you focus on other high-value work.

Now just wait as Claude delivers to you a well-crafted response that effectively accomplishes your task. But the real power of Claude comes with continued and frequent communication, not just one-off prompts. The more you interact with Claude, the better it becomes at understanding your communication style and preferences for future conversations.

Working with Claude, you have a great partner and a powerful intelligent collaborator that amplifies your capabilities across everything you work on. Claude brings AI intelligence to your work. You bring the context and expertise that makes the work meaningful.