Scribd
Upload
179 views1 page

Prompt Engineering

The document outlines various roadmaps related to AI and data science, including AI Engineer, AI and Data Scientist, and MLOps. It covers key concepts in prompt engineering, including techniques, best practices, and considerations for using large language models (LLMs). Additionally, it emphasizes the importance of continuous learning and provides links to interactive versions of the roadmaps.

Uploaded by

kingsleydesigns37
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
179 views1 page

Prompt Engineering

The document outlines various roadmaps related to AI and data science, including AI Engineer, AI and Data Scientist, and MLOps. It covers key concepts in prompt engineering, including techniques, best practices, and considerations for using large language models (LLMs). Additionally, it emphasizes the importance of continuous learning and provides links to interactive versions of the roadmaps.

Uploaded by

kingsleydesigns37
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

Related Roadmaps

Find the interactive version of


AI Engineer this roadmap and more at

AI and Data Scientist


MLOps Roadmap
Prompt Engineering roadmap.sh

AI Red Teaming
Prompt Engineering

LLMs and how they work?

Common Terminology Introduction What is a Prompt?

What is Prompt Engineering?


LLM Tokens

Context Window Models offered by ____


LLM Configuration
Hallucination Agents OpenAI Google

Prompt Injection Anthropic Meta

Model Weights / Parameters Sampling Parameters xAI

Fine-Tuning vs Prompt Engg. Temperature

AI vs AGI RAG Top-K Top-P


Output Control

Max Tokens

Stop Sequences

Prompting Techniques Structured Outputs Repetition Penalties

Frequency Penalty

Presence Penalty

Zero-Shot Prompting Step-back Prompting

One-Shot / Few-Shot Prompting


Chain of Thought (CoT) Prompting Automatic Prompt Engineering

System / Role / Contextual


Use LLM to generate Prompts
Self-Consistency Prompting
System Prompting

Role Prompting Tree of Thoughts (ToT) Prompting

Contextual Prompting ReAct Prompting

Prompting Best Practices AI Red Teaming

AI Red Teaming Roadmap

— Provide few-shot examples for structure or output style you need

— Keep your prompts short and concise

— Ask for structured output if it helps e.g. JSON, XML, Markdown, CSV etc

— Use variables / placeholders in your prompts for easier configuration

— Prioritize giving clearer instructions over adding constraints

— Control the maximum output length


Improving Reliability
— Experiment with input formats and writing styles
Prompt Debiasing
— Tune sampling (temperature, top-k, top-p) for determinism vs creativity

— Guard against prompt injection; sanitize user text Prompt Ensembling

— Automate evaluation; integrate unit tests for outputs


LLM Self Evaluation
— Document and track prompt versions

— Optimize for latency & cost in production pipelines Calibrating LLMs

— Document decisions, failures, and learnings for future devs

— Delimit different sections with triple backticks or XML tags

Continue learning with following roadmaps

AI Engineer Roadmap AI Agents Roadmap

You might also like