Most people who try AI start with ChatGPT. That's sensible. It's the most widely known tool, the free tier is genuinely useful, and there are more tutorials for it than anything else. If you completed Module 1 using ChatGPT, that's exactly right.
But ChatGPT is not the only option, and for certain tasks it's not the best one. This module is about Claude, the AI assistant built by Anthropic. It's genuinely different from ChatGPT in a few ways that matter for beginners, and there are specific situations where it's the better call.
This is not an endorsement. Claude has weaknesses too, and we'll cover those. The goal here is to give you an honest read on when switching tools will actually improve your results.
What Claude is particularly good at
Thinking through a problem with you
If you have a complex question, a career decision to work through, a situation with multiple moving parts, something where you want to think out loud rather than get a direct answer, Claude tends to do this better than the base ChatGPT.
It's more likely to ask clarifying questions before jumping to conclusions. It tends to lay out the relevant considerations before rendering a verdict. It handles "I don't know what to do here, help me think" better than "give me the answer."
For career-shifters specifically, this matters. Career decisions are not clean problems with optimal solutions. They involve trade-offs between security and interest, geography, family, financial timelines, risk tolerance, and things you can't fully articulate yet. An AI that helps you structure your own thinking is more useful for this than one that produces a confident five-step plan.
A real beginner use case: Drop a decision you're wrestling with into Claude. Not "what should I do?" but "Here's the situation: [full context]. I'm trying to work out the right move. What questions should I be asking myself that I'm not asking yet?"
Reading and summarising long documents
This is Claude's most consistently impressive capability for beginners.
Claude has a very large context window, meaning it can process long documents in a single conversation. You can paste in a 40-page research paper, a contract, a set of course notes, an entire government report, or a lengthy article and ask Claude to explain it, summarise it, pull out the most important points, or help you understand the parts that confused you.
This is enormously useful if you're studying a new field. One of the hardest parts of entering a new domain is not finding the information, it's understanding it when you can't yet recognise what matters. Claude can help bridge that.
Specific to career-shifters: Find the most important textbook, white paper, or definitive report in the field you're moving into. Paste large sections into Claude. Ask: "What are the 5 to 10 concepts I'd need to understand to follow a professional conversation in this field?" You'll get a genuine curriculum, grounded in the actual content of the field, rather than a generic "beginner's guide" AI hallucination.
Writing in your specific voice
Claude tends to be better than the base ChatGPT at mimicking a writing style when given examples to work from. If you paste in three emails you've written, three LinkedIn posts you've published, or three pieces of writing that reflect how you communicate, and then ask Claude to write something new in that voice, the result is typically more accurate to your actual style.
This matters for cover letters, LinkedIn content, professional communications, anything where "this sounds like me" is important.
The technique: use Pattern 2 from Module 2 (examples-first). Give Claude three pieces of your actual writing before asking it to write anything new. The voice match will be noticeably better.
Structured reasoning and analysis
If you need to compare options, evaluate trade-offs, or build a structured argument, Claude tends to produce cleaner output than models that reach for bullet points and lists by default. Its prose reasoning is often more useful for actual thinking than a formatted list.
For example: ask Claude to evaluate whether a specific further education programme is worth doing given your specific career goals and financial situation. Give it all the context. It will walk through the relevant factors in a way that reads like a thoughtful colleague who has processed the information, not a list of pros and cons you could have made yourself.
The real beginner use case: CV critique against a job ad
Here is the most immediately useful thing a career-shifter can do with Claude in the next 10 minutes.
Find a job ad for a role you're genuinely interested in. Paste it into Claude alongside your current CV. Then ask this:
Here is a job ad for a role I'm applying for:
[paste job ad]
Here is my current CV:
[paste CV]
I want you to do three things.
1. Tell me honestly where my CV matches the requirements well, and
where there are obvious gaps.
2. Tell me which specific parts of the job ad I should address in my
cover letter, because my CV doesn't clearly answer them.
3. Suggest two or three ways I could reframe existing experience in my
CV to be more relevant to this role, without misrepresenting anything.This prompt, with your real documents, will produce something genuinely useful. Not generic career advice. Specific, grounded feedback on your specific application.
Claude's honest weaknesses
No web access by default. The standard Claude interface does not browse the internet. If you ask it what happened last week, or about current job market data, it will either not know or will tell you its knowledge has a cutoff. Gemini (Module 6) is the better call for anything requiring current information.
Images on the free tier. The Claude.ai free tier has limited image capabilities. If you need to upload an image for analysis (a photo of a textbook diagram, a screenshot you want explained), ChatGPT's free tier currently handles this better.
Speed. Claude can occasionally be slower to respond than ChatGPT, especially on complex requests. This is a minor inconvenience, not a dealbreaker.
Occasional over-caution. Claude is built by a company (Anthropic) that is particularly focused on AI safety, which means it sometimes declines to help with things that are genuinely harmless. If this happens and the request is legitimate, trying a different framing usually resolves it.
Where to find Claude
Claude is at claude.ai. Free account required. The free tier is Claude Sonnet, which is genuinely capable. The paid tier (Claude Pro) gives you access to Claude Opus, which is the more capable model, but for the tasks in this course, Sonnet is sufficient.
One honest note: this course was built using Claude as a drafting assistant. That felt appropriate. Teaching AI honestly means using it, not pretending you didn't.