2 answers
2
Updated 91 views

How can I integrate AI as a tool with my school life and not like a crutch? Something where it's easier with it but I can work without it? #Spring26

I am a CS major and just would like to learn how to better automate some mundane tasks so I can focus more on the important stuff.

spring26 programming coding timemanagement studytips academicsuccess artificialintelligence automation computersciencemajor digitaltools
Answer This
2

2 answers

0
Updated
Share a link to this answer
Share a link to this answer

Peter’s Answer

For learning, you should treat the AI as a partner, as a tutor and as a collaborator. Ask it to explain things to you better when you don't understand them, this goes for code and for your assignments and lectures. Think critically, ask it for references, or to search or verify facts to confirm them. For coding, use AI to ideate, use a smart model to help you plan, debug, and test, because systems engineering, good taste, and testing and verification are now more important than they ever were before. AI will make mistakes, your job now is to learn enough and be smart enough to catch those mistakes and fix them by thinking critically and by using and testing the systems you build collaboratively with AI. In general try to build systems that are as simple as possible while still being useful.

Peter recommends the following next steps:

https://chatgpt.com/features/study-mode/
https://cs50.harvard.edu/extension/2024/fall/notes/ai/
https://teachingcommons.stanford.edu/teaching-guides/artificial-intelligence-teaching-guide/understanding-ai-literacy
0
0
Updated
Share a link to this answer
Share a link to this answer

Liam’s Answer

I am going to take your question a few steps back from actually using the AI for tasks. Build a home lab, host local AI. Start with that because likely you own a computer that can run some lightweight low token LLMs and start to help you understand how you can implement them for yourself. Start with Ollama or LM Studio and get some coding models from huggingface. Set up a small lab for yourself and figure out how to get it to code and automate small tasks for you. See what you can get it to do, see what your limitations are. Remember this is a LAB and it isn't PROD so don't put valuable data on your machine!!

Make some agents and see what they can help you with. Figure out what kind of models work on what kind of tasks. Start your prompt library so you can try those prompts out on more powerful models later on!

I recommend anyone in STEM to make a homelab because you really get the control and experimentation you need to understand systems at massive scale. You may not be able to afford subscriptions to APIs for larger providers, this is a way around that. If you can make a prompt or agent work on a really scaled down LLM, you will definitely be able to get it to work on massive scale models later.

Skip OpenClaw for now unless there is something really specific you need it for.

Liam recommends the following next steps:

https://www.youtube.com/live/TsyAsrnYnhQ
https://youtu.be/GrLpdfhTwLg
https://youtu.be/budTmdQfXYU , https://youtu.be/Wjrdr0NU4Sk
https://ollama.com/library/qwen2.5-coder - note you can run this as a 7b model (like on an ok gaming machine) and in claude code all locally
0