Prompt Engineering For Developers - In a Nutshell

Ashutosh Verma — Thu, 15 Jan 2026 05:31:18 +0000

Two type of LLM

Base LLM (Pre-trained LLMs) - Base LLMs undergo pre-training on massive unlabeled text datasets to predict the next word or token, gaining broad language understanding and knowledge. These models excel at tasks like text completion or generation but often respond loosely to prompts, producing generalized or unpredictable outputs without specific guidance.
Instruction Tuned LLM - Instruction-tuned LLMs start from a base model and undergo supervised fine-tuning on datasets of instruction-response pairs, teaching them to follow explicit user directives like “summarize” or “explain.” This process, often enhanced by reinforcement learning from human feedback (RLHF), improves precision, reduces hallucinations, and enables better handling of diverse tasks such as question-answering or step-by-step reasoning.

Guidelines

Principle-1: Write clear and specific instructions

Use delimiters - such as triple quotes, backticks, angle brackets etc. to define what are instructions and what is the actual text on which task is to be performed in your prompt
Ask for structured output
Check whether conditions are satisfied, check assumptions are required to do the task
Few-shot prompting - Give successful examples of completing tasks, then ask model to perform the task

Principle-2: Give model time to think

1. Specify the steps to complete a task

Step-1: ______

Step-2: ______

……

Step-N: ______

2. Instruct the model to workout its own solution before rushing to a conclusion

Ask model to solve the problem itself first, then compare with your given solution.

For example, to solve a maths problem give the question first then ask model to solve the question first and also give your solution for comparison in your prompt

Model Limitations

LLM might not have memorized the information it has seen while training. So sometimes it gives plausible response but not actually true
Hallucinations → Make statements that sound plausible but not true

Iterative Prompt Development

Source: deeplearning.ai

Prompt guidelines

Be clear and specific
Analyze why result does not give desired output
Refine the idea and the prompt
Repeat

→ Iteratively refine the prompt for the length and style

Iterative Process

Try something
Analyze where the result does not give what you want
Clarify instructions, give more time to think
Refine prompts with a batch of examples

LLM Capabilities

Summarizing
Inferring → Identifying sentiments, identifying emotions, identifying anger, Extracting something, inferring topics from a text
Transforming → Translation, Tone transformation, Format conversion (e.g. - JSON to HTML), Spellcheck/grammar check
Expanding → e.g. Customize the automated reply to a customer email
Chatbot → Setting Role: system/assitant/user while chatting with the model

Source: deeplearning.ai

Reference: https://www.deeplearning.ai/short-courses/chatgpt-prompt-engineering-for-developers/

How I Prepared for the AWS Cloud Practitioner Exam

Ashutosh Verma — Mon, 15 Sep 2025 18:11:51 +0000

Amazon Web Services (AWS) has always fascinated me with its brilliantly designed and impeccably organized cloud ecosystem. Every time I explore AWS, I am struck by how seamlessly all its components come together to create a reliable, scalable, and innovative platform. Its thoughtful architecture and vast array of services never cease to inspire awe, revealing the power and elegance behind one of the most transformative technologies of our time.

For a long time, I have aspired to earn an AWS certification, but my limited knowledge held me back. Determined to change that, I began researching AWS certifications to understand which path would best suit my beginner level. That’s when I discovered the AWS Certified Cloud Practitioner exam—a perfect starting point for newcomers like me seeking to build a strong foundation in cloud computing.

While exploring other people’s experiences with the AWS Certified Cloud Practitioner exam, I came across several highly recommended instructors, and one name that stood out was Stephane Maarek. Curious, I read up on him and his comprehensive course. Then, one day while browsing Udemy, I saw his course, bundled with six practice test papers, offered at a discounted price. Without hesitation or second thoughts about how I would find the time to study, I seized the opportunity and purchased the course.

I began the course in June, immersing myself fully and taking handwritten notes—which I love—because writing helps me retain information better. To help with last-minute revision, I created a mind map on Miro, allowing me to quickly review and connect key concepts before the exam. After completing 4 to 5 sections, I went on a week-long trip, and upon returning, a busy spell at work made it challenging to find study time. Despite these interruptions, I resumed the course in August and was proud to complete it within the same month.

Eager to take the exam as soon as possible, I crafted a strategic study plan. The course consisted of around 15 sections, so I grouped them into clusters of 2-3 based on their length and content. Since I could only study after office hours with about 2-3 hours available each day, I balanced each group thoughtfully—pairing a section that was easier to digest with one that was more challenging, or mixing shorter sections with longer ones. This approach helped me manage my study sessions effectively and keep the momentum going.

In addition to covering the course material, I made it a point to find some free moments at work to review practice question dumps available on Github. This helped me gain a clearer understanding of the exam pattern and types of questions. Whenever I encountered new or challenging topics, I took the initiative to research them further online, deepening my grasp of the subject.

In the second week, I scheduled my exam for Sunday and committed to tackling all six practice papers—one each day of the week. I took each test in strict exam mode, completing it within the allotted 1.5 hours. After submitting each paper, I carefully reviewed my mistakes and studied the provided explanations. For every AWS service I encountered, I made detailed notes and researched a high-level overview online, using tools like Perplexity to deepen my understanding.  On Sunday, I took the exam at the testing center, passed, and received my certification on the same day.

Tips for Preparing for the AWS Certified Cloud Practitioner Exam

1. Get Hands-On Experience
Practice with popular and useful AWS services like EC2, Lambda, CloudWatch, CloudTrail, ECS, Route 53, and others. Hands-on experience helps reinforce concepts better than just reading.

2. Focus on a High-Level Overview
Avoid getting overwhelmed by diving too deep into numerous white papers and documentation. Aim to understand the core concepts and purpose of each service since the exam tests your overall understanding. Detailed reading can come later when working on real projects.

3. Follow Stephane Maarek Course 
The course by instructor Stephane Maarek is comprehensive and covers almost all topics needed for the exam. His structured approach is highly recommended.

4. Review Official Exam Guidelines 
After finishing the course, check the official AWS Certified Cloud Practitioner exam page and carefully read the in-scope services section to ensure you’re aligned with the exam requirements.

5. Associate Keyword Definitions
Link every service with a concise keyword or definition to aid memorization, for example:

AWS Config: record configuration of resources
AWS Organizations: multi-account management
CloudFormation: infrastructure automation with YAML templates
Amazon Neptune: Graph database

6. Learn From Others’ Experiences
Read blogs and experiences shared by other exam takers to explore various strategies. Choose the one that fits your learning style and personality best.