AI Programming for Beginners: Essential Tools and Productivity Tips

For those with no prior exposure to AI or programming, starting from scratch can feel completely overwhelming. This beginner's guide comes from a Bilibili creator who teaches AI programming — and the core message is clear: instead of agonizing over which language to learn first, just start using AI tools and let top-tier AI assistants become your personal tutors along the way.

Before Learning AI Programming, Get Your Mindset Right

Many beginners run into frustrating experiences their first time with AI — responses that miss the point, hallucinations, inaccurate answers. It can feel like AI is more "artificial" than "intelligent." The creator emphasizes that this is a normal part of AI's development, and it's no reason to give up.

The key is more exposure, more interaction. Every AI model has its own "personality" and the types of problems it handles best. Only by genuinely spending time with AI will you start to understand its strengths and weaknesses — and know which tool to reach for when working on real projects.

The advice is straightforward: take every question you'd normally search online, and ask AI instead. This dramatically increases your time with these tools, and that time is the single most important variable in developing an intuitive feel for them.

Errors and Bugs Are Part of the Process

When learning to code, the environment shown in a tutorial and your actual setup will often differ. Running into errors while installing tools or configuring environments is practically inevitable. The creator specifically notes: don't treat these small bugs as roadblocks, and don't get frustrated just because something worked in the video but not on your machine.

What matters more is changing how you approach problems — when you hit an error, the first step is to ask AI. The creator even makes a bold claim: AI's programming ability surpasses any individual human on earth. It's like having a world-class developer on call, ready to serve and explain at any moment.

Start with Doubao: Your First AI Tool

For complete beginners, the creator recommends starting with Doubao as your first tool — it's widely used and easy to get into.

The onboarding path is simple: open the web version of Doubao (just search for it), sign in with your phone number, and get familiar with the interface. At this stage, don't overthink prompt engineering — just build the habit of turning to AI whenever you have a question.

How to Ask AI the Right Way

The creator shares a practical framework for asking questions, especially useful for beginners:

• Explain your situation: Tell the AI what you're trying to do and what kind of computer you're using.
• Paste the error directly: Send the exact error message or problem as-is.
• Identify yourself: Explicitly tell the AI "I'm a programming beginner" — and when you encounter unfamiliar terms, ask it to explain them again.

The clearer you are about your context, the better AI can calibrate its response to your level. And unlike any human teacher, AI has infinite patience — keep asking follow-up questions and it will keep explaining until you get it. This practice of follow-up questions + refining your prompts is the core skill for getting genuinely useful answers from AI.

Web, Desktop, Mobile: Use All Three

Once you're comfortable with the web version, the creator suggests also installing Doubao's desktop and mobile apps — each version has its strengths.

The desktop app offers more detailed features worth exploring one by one. It even supports having AI "watch" as you install software and guide you through each step — incredibly helpful for beginners troubleshooting real issues on their machines.

The mobile app shines with voice conversation — if your computer has a microphone, you can use voice chat there too. The creator encourages beginners to voice their questions out loud: "it's exactly like talking to a person." The more you talk, the better you'll understand what AI is and isn't good at answering.

Expanding from One Model to Many

The creator repeatedly drives home one principle: don't install every AI tool at once. Focus on mastering one first (Doubao), then gradually branch out.

Once you're comfortable with Doubao, try other domestic models like Alibaba's Tongyi Qianwen or Moonshot AI's Kimi — both available on web, desktop, and mobile. Different models have different capabilities, and using several helps you develop the comparative judgment to know which to use when.

For those with access, international models are also worth exploring: the paid ChatGPT is the most well-known, while Google's Gemini offers a solid free option. These are more advanced choices though — beginners can absolutely start with Doubao alone.

A quick overview of these tools helps with later decisions. Tongyi Qianwen is Alibaba's large language model, known for stable Chinese comprehension, long-document handling, and integrated code generation and image recognition. Kimi, developed by Moonshot AI, gained early recognition for its ultra-long context support — capable of ingesting hundreds of thousands of characters at once — making it well-suited for lengthy reports or large codebases. ChatGPT from OpenAI, particularly the GPT-4 series, has a strong reputation for logical reasoning and programming assistance, though it requires a paid subscription and can be tricky to access outside certain regions. Gemini is Google's multimodal model — the free tier is usable and it excels at incorporating real-time information from Google Search. Understanding these basic differences will help you, once you've built up some experience, match the right tool to the right task.

The Real Goal: Accumulating Time with AI

The article closes by returning to its through-line: truly learning to use AI well in your work and projects requires long-term, sustained exposure. The creator offers a concrete benchmark — roughly 2,000 to 3,000 hours of continued engagement.

Only through that kind of accumulation will you naturally develop the judgment to know: which model to use, how to prompt it, and how to work through problems with AI's help. For beginners, the most solid first step into AI is simply to actually start using the tools — beginning with Doubao, running every question from your studies, daily life, and projects through AI.

The "2,000 to 3,000 hours" benchmark echoes the spirit of the well-known "10,000-hour rule" from psychology — both emphasize deliberate practice and sustained engagement over passive accumulation. With AI tools, the real substance of this time is a feedback loop: encountering real problems, forming questions, evaluating response quality, and refining how you ask. Every follow-up question, every time you catch an AI mistake and correct course, builds your intuitive sense of where a model's capabilities end. That kind of intuition can't be acquired by reading tutorials or watching videos — it only develops through real use. So rather than thinking of "3,000 hours" as a threshold to clear, think of it as a signal: start accumulating from day one, and every time you use AI to solve a real problem, you're moving forward.