Long time not writing, lets start

Introduction

It’s been a while since I last wrote a blog post, and I’m excited to start again. With the new year approaching, I want to revive my old habit of writing and share my thoughts with you. In this post, I’ll be writing randomly without a structure, so bear with me.

My Goals for 2025

I’ve created a list of easy-to-do habits that I want to maintain consistently. These include:

• Attending work and being punctual
• Morning jogs
• Reading a book for at least 30 minutes a day
• Practicing Duolingo for 15 minutes a day
• Writing a blog post daily
• Doing abs exercises to get those elusive six-packs
• Taking care of my skin before bed

I’ve already completed 4 of these tasks today, and I’m feeling motivated to continue.

My Resolutions for 2024 and 2025

In 2024, my resolutions were:

• Save $100 million (achieved)
• Get a girlfriend
• Earn a double-digit salary per month
• Get six-packs and weigh around 68 kg with under 20% body fat

For 2025, I’ve added some new resolutions:

• Save $100 million (again, because it’s a great goal)
• Get a girlfriend
• Learn the basics of Rust programming language
• Learn Dutch or Japanese
• Get six-packs and weigh around 68 kg with under 20% body fat

Daily Sharing

I want to share something interesting with you every day. Today, I’ll be talking about Prompt Engineering.

What is Prompt Engineering?

Prompt Engineering is the process of designing and refining inputs or questions to improve the quality, relevance, and accuracy of AI responses. It’s like teaching a language model to understand what you want it to say.

What is a Prompt?

A prompt is an input or question given to a language model to guide its response output. The goal is to shape the output of the AI model by specifying the information or type of response the user is seeking.

5 Common Mistakes in Prompting and How to Fix Them

1. Vague Direction: Give the AI clear direction on what you want it to do.
2. Unformatted Output: Specify the format you want the AI to use.
3. Missing Example: Provide examples to guide the AI’s response.
4. Limited Evaluation: Evaluate the quality of the AI’s response.
5. No Task Division: Divide tasks into smaller, manageable parts.

Advanced Techniques

• Few-shot Prompts: Provide the AI with examples to help it understand the desired response style.
• Chain-of-Thought (CoT) Prompting: Encourage the AI to break down its reasoning into steps.
• Self-Consistency: Generate multiple outputs and select the most consistent response.
• Knowledge Generation Prompting: Prompt the AI to generate relevant background knowledge before answering a question.
• ReAct: Combine reasoning with action-oriented responses.

These techniques can help you get the most out of language models like ChatGPT, Llama, and others.

Conclusion

That’s it for today’s post. I hope you found it interesting and informative. I’ll be back with more content soon. Thanks for reading!

If I had an hour to solve a problem, I’d spend 55 minutes thinking about the problem and five minutes thinking about solutions

– Albert Einstein