A Journey through Diverse Patterns and Future Horizons in Generative AI: Prompt Engineering represents a rapidly evolving field within AI. The landscape of prompt patterns has become extensively documented and diversified, encompassing LLM reasoning approaches that span from one-shot to Chain-of-Thought and Tree-of-Thought prompting. In this session, our focus will delve into the evolutionary journey and the future prospects of prompt engineering. We will draw parallels with the historical progression of programming patterns and technologies, offering valuable insights and cross-domain learning opportunities. Join us to explore the dynamic realm of prompts and their role in shaping the landscape of Generative AI.
In the evolving digital landscape, recommendation systems are a critical component in driving customer engagement and satisfaction. With the advent of large language models (LLMs) like GPT-4, a new vista of possibilities has opened up, extending the capabilities of these systems. This session will delve into the intriguing confluence of LLMs and recommendation systems, revealing the future of personalized experiences.
The session will kick off with an introduction to LLMs, followed by an examination of their synergy with recommendation systems. We'll explore the advantages LLMs bring, such as superior context understanding, predictive accuracy, and rich, nuanced personalization. We'll also discuss the challenges, such as potential biases and resource requirements.
A highlight will be case studies, showcasing real-world applications of LLMs in recommendation systems and illustrating how businesses are leveraging this technology to revolutionize their customer engagement strategies.
Embark on a journey to the Universal Creativity Era, where Generative AI supercharges imagination across design domains. Delve into real-world examples from diverse sectors and with brands such as Disney, Marvel, Starbucks...
This session showcases AI's transformative role in many different design workflows, from ad campaigns, product and character design, video production, movie casting and costume design, co-branding activities, and even theme park menus design!
Witness how AI democratizes creativity, blurring traditional boundaries and welcoming everyone to leave their unique mark in this grand innovation tapestry.
Generating non-existing city concepts, materials and architectural design.
Prompt Engineering is revolutionizing business scenarios. But how about the Intellectual Propoerty (IP) rights of this immersive and impactful activity? Is there any IP? Who does it belong to? How to secure it? With rights come duties, what are they?
In the rapidly-evolving landscape of AI, the way we interact with generative models significantly impacts their usefulness and the outcomes they produce. This session aims to shed light on a technique to streamline AI interactions: pseudocode prompting.
Natural language is great for simple prompts, but it often lacks the precision required for more complex or algorithmic interactions. Enter pseudocode, a simplified semi-formal language that can efficiently guide AI towards desired responses while following specific rules.
Through this session, we'll discuss:
* Principles of pseudocode prompting and how it improves reasoning performance over natural language prompts.
* Practical strategies and methods for generating and optimizing pseudocode prompts for various applications.
* How pseudocode prompts can reduce token usage by 20-30%, leading to decreased costs and faster responses.
* How pseudocode can democratize programming, making it accessible to a wider audience by allowing the specification of requirements as natural-language rules or constraints.
* Illustrate how structured pseudocode offers easier navigation and maintenance of complex prompts compared to natural language through scope blocks, indentation, and visual encapsulation.
* Enhancing Response Quality: Using structured templates and queries to minimize poorly-formed responses and reduce token usage.
This session will focus on the practical application of pseudocode in AI, making it highly suitable for those interested in deploying more efficient and robust AI applications.
Whether you are a seasoned data scientist, a developer looking for innovative ways to interact with AI models, or simply an AI enthusiast, this session will equip you with fresh insights and practical strategies for a more efficient AI prompting experience.
What fine-tuning is, how it works, its benefits for prompt engineers and production deployments, and how to migrate from an instruction-based prompt to a fine-tuned LLM.
🚀 Get ready to supercharge your development with Semantic Kernel and Semantic Kernel Tools! This powerful SDK and extension for Visual Studio Code lets you mix conventional programming languages with the latest in Large Language Model AI to deliver delightful and intelligent experiences.
With SK, you can quickly and easily integrate AI capabilities into your applications without needing to train or fine-tune a model from scratch. And with SK's Skills, Memories, and Connectors, you'll have maximum flexibility to build new experiences that bring unparalleled productivity for your users.
Plus, with Semantic Kernel Tools, you'll be able to develop your own semantic skills faster and with greater ease. So, what are you waiting for? Join the AI revolution with Semantic Kernel and Semantic Kernel Tools today! 🤖
As a Product Manager for AI tools, I see a lot of challenges arising when the uncertainties of AI world meet the well-defined processes of classical engineering projects. These challenges include:
- The nondeterministic nature of natural language inputs and outputs, that creates a lot of uncertainty and widens the gap between the working Proof-of-concept and a well-rounded enterprise level solution.
- Roles shift: the prompt engineering and testing effort highly relies on team members without engineering background (content writers, marketing professionals) who are experts in language but may lack familiarity with R &D processes.
- Speed vs. Quality: this old as time conflict creates more tension than ever when a company needs to adopt AI fast without compromising the high quality and security standards
All these things create a lot of friction in big projects, where previously there was a smooth and clear way of planning, execution, testing and delivery.
In this talk, I'll suggest a few ways to optimize the work, to build good prompt evaluation processes, to teach project teams about prompt engineering and to build a process in which prompt engineering is a normal, well defined and somewhat predictable part of the workflow
ChatGPT can help you brainstorm. GitHub's Copilot can be a great rubber duck for your codebase. But, what if you want to build something new?
In this talk we'll explore how to leverage current tooling to explore open source models, quickly spin up a model with a persona and instructions, and even build out an MVP with the help of a bot empowered with the ability to write and execute code.
You'll learn about some exciting new tools as well as some tips and tricks for prototyping your next idea using language models.
In today's fast-paced development cycles, product managers are often swamped with the task of continuously reviewing project tickets to assess the status and potential risks. Our presentation dives into how prompt engineering can be a game-changer in this process. Utilizing natural language processing, we showcase a method for parsing and interpreting the vast array of project tickets to instantly categorize them based on urgency, risk, and other vital metrics originated from internal hackathon project we did to help leadership understand the risks of project from reading thousands of tickets in one pass using LLM models. This enables product managers to make well-informed decisions more efficiently. Furthermore, we present a prototype of an AI-powered assistant built on this methodology, designed to assist product managers in real-time project evaluation and risk management.
Explore the frontier of data visualization in this session, where we merge the prowess of generative AI with the elegance of Vega-Lite in Microsoft Power BI. Witness the power of Iterative Prompt Engineering as we guide ChatGPT to craft intricate visuals from raw data. Delve into this innovative technique and discover a novel way to tell compelling data stories. Join us for a transformative look at the future of AI-driven data visualization.
In this session, we will delve into the practical and effective use of Prompt Engineering in designing detailed, accurate, and user-friendly OpenAPI descriptions for APIs. OpenAPI, a widely accepted standard for describing RESTful APIs, plays a critical role in API design, development, and usage. It allows both humans and machines to understand and interact with the capabilities of a service without direct access to the source code.
However, creating accurate and comprehensive OpenAPI descriptions can be a daunting task. This is where Prompt Engineering, a cutting-edge technique, can significantly enhance the process. It enables AI models to understand and generate more accurate, contextually relevant responses, hence perfect for crafting intricate API descriptions.
In this session, we will explore strategies for translating API operations into natural language prompts, effective ways of refining API models based on generated prompts, and practical demonstrations of how these techniques can improve the API development workflow. Attendees will gain a deeper understanding of how to leverage AI and Prompt Engineering in producing high-quality OpenAPI descriptions, leading to better-designed, more robust, and easily understandable APIs.
In a world where generative AI is reshaping the boundaries of what's possible, the power of prompting emerges as the game-changer.
But can it truly be a seismic catalyst for those with a growth mindset?
This session provides the answer by delving deeper into the transformative journey of harnessing generative AI, not just as a tool, but as a partner in exploration, personal development, and growth.
Dive into a personal journey of rediscovery and innovation as we explore 'Mystical Fusion' Tarot system, not as an end product, but as a testament to the creative learning process: the challenges faced, the boundaries pushed, and the lessons learned.
From revisiting a decades-old idea with limited to no programming experience to the triumph of creating a tangible delivery of the project, it's a story of resilience, innovation, and the boundless possibilities that emerge when a growth mindset meets the right tools.
By the end of this journey, what was once a hypothesis before the experiment is now proven, culminating in a fully functional AI-powered Tarot app running on its own server in the cloud, a physical deck of 52 cards, a guiding e-book, and a website, all created during a 2 and a half week creative holiday break.
The proof has been delivered. The power of prompting is rocket fuel for people with a growth mindset. The only limitation is your own creativity and energy, or like Charlie Chaplin once said: "Imagination is nothing without doing".
At numerous conferences, there is a growing discussion about the impressive capabilities of the OpenAI API, ChatGPT, MidJourney, and other similar tools. However, a common question consistently arises among the audience: "How can we effectively write prompts?" - The most important task working with those models. Let's answer that question!
Sending a message to ChatGPT or other AI tools seems simple, but expertise lies in mastering prompts for desired outputs. Talks usually cover applications, model explanations, or API usage. This talk is different! In this talk we'll dive into Prompt Engineering!
During this session, Jeroen will explain what Prompt Engineering is, why it is important to learn, will guide you through a step-by-step process called the Crafting AI Prompts Framework, which he has personally developed to create optimal prompts. By the end of the presentation, you will be equipped to effectively utilize the framework, simplifying text-based tasks such as content writing and review, while also expanding into areas like image and slide deck creation, user story development, and even coding: only by using the right prompts!
Don't miss out on this opportunity to enhance your skills, boost your productivity, and stay ahead of the curve!
I will go through some simple steps that can help you improve how you debug issues in PowerShell. The methods can be applied to. other languages also. Can it also help you code more effectively?
Experience how natural language can assist you and your management team in developing comprehensive strategies based on OKR (Objectives and Key Results) principles, as well as setting up supporting initiatives and projects.
The live demonstration will utilize Azure OpenAI, ChatGPT, and the Microsoft Copilot ecosystem.
The speaker possesses more than 20 years of experience in project portfolio management and has been a Microsoft MVP for 12 years.