Rong Lu

𝐀𝐠𝐞𝐧𝐭𝐢𝐜 𝐖𝐨𝐫𝐤𝐟𝐥𝐨𝐰: 𝐏𝐢𝐨𝐧𝐞𝐞𝐫𝐢𝐧𝐠 𝐭𝐡𝐞 𝐅𝐮𝐭𝐮𝐫𝐞 Recently I happened to see this nice video ( 𝘩𝘵𝘵𝘱𝘴://𝘸𝘸𝘸.𝘺𝘰𝘶𝘵𝘶𝘣𝘦.𝘤𝘰𝘮/𝘸𝘢𝘵𝘤𝘩?𝘷=𝘴𝘢𝘭78𝘈𝘊𝘵𝘎𝘛𝘤 ) by Andrew Ng in which he talks about Agentic workflows. In Language Model (LLM) development, we provide a zero-shot prompt and receive an answer. We may review it and add more prompts to get it to a desired shape. This is exactly what the agentic workflow does. Basically, 𝘢 𝒃𝒖𝒏𝒄𝒉 𝒐𝒇 𝑨𝑰 𝒂𝒈𝒆𝒏𝒕𝒔 𝒘𝒐𝒓𝒌 𝒕𝒐𝒈𝒆𝒕𝒉𝒆𝒓 𝒊𝒏 𝒂𝒏 𝒊𝒕𝒆𝒓𝒂𝒕𝒊𝒗𝒆 𝒘𝒂𝒚 𝒕𝒐 𝒑𝒓𝒐𝒅𝒖𝒄𝒆 𝒂 𝒅𝒆𝒔𝒊𝒓𝒆𝒅 𝒐𝒖𝒕𝒑𝒖𝒕. Moreover, he even claims the output produced by agentic workflow is much better than the output produced single LLM which might sound obvious but what surprised me was, as per the “𝑯𝒖𝒎𝒂𝒏𝑬𝒗𝒂𝒍” coding benchmark, agents-based approach beating plain LLM by quite a margin (see the picture below). The plain 𝐆𝐏𝐓 3.5 was at 48% but 𝑳𝑫𝑩+𝑹𝒆𝒇𝒍𝒆𝒄𝒕𝒊𝒐𝒏 (𝘈 𝘓𝘢𝘳𝘨𝘦 𝘓𝘢𝘯𝘨𝘶𝘢𝘨𝘦 𝘔𝘰𝘥𝘦𝘭 𝘋𝘦𝘣𝘶𝘨𝘨𝘦𝘳 𝘷𝘪𝘢 𝘝𝘦𝘳𝘪𝘧𝘺𝘪𝘯𝘨 𝘙𝘶𝘯𝘵𝘪𝘮𝘦 𝘌𝘹𝘦𝘤𝘶𝘵𝘪𝘰𝘯 𝘚𝘵𝘦𝘱-𝘣𝘺-𝘴𝘵𝘦𝘱) fared at impressive 95.1% with the same GPT 3.5. The plain 𝐆𝐏𝐓 4 was at 67% but 𝑨𝒈𝒆𝒏𝒕𝑪𝒐𝒅𝒆𝒓 (𝘔𝘶𝘭𝘵𝘪-𝘈𝘨𝘦𝘯𝘵-𝘣𝘢𝘴𝘦𝘥 𝘊𝘰𝘥𝘦 𝘎𝘦𝘯𝘦𝘳𝘢𝘵𝘪𝘰𝘯 𝘸𝘪𝘵𝘩 𝘐𝘵𝘦𝘳𝘢𝘵𝘪𝘷𝘦 𝘛𝘦𝘴𝘵𝘪𝘯𝘨 𝘢𝘯𝘥 𝘖𝘱𝘵𝘪𝘮𝘪𝘴𝘢𝘵𝘪𝘰𝘯) fared at whooping 96.3% with the same GPT 4.  You can get more information about the evaluation and the models at 𝘩𝘵𝘵𝘱𝘴://𝘱𝘢𝘱𝘦𝘳𝘴𝘸𝘪𝘵𝘩𝘤𝘰𝘥𝘦.𝘤𝘰𝘮/𝘴𝘰𝘵𝘢/𝘤𝘰𝘥𝘦-𝘨𝘦𝘯𝘦𝘳𝘢𝘵𝘪𝘰𝘯-𝘰𝘯-𝘩𝘶𝘮𝘢𝘯𝘦𝘷𝘢𝘭. He also mentioned the following are the 4 design patterns that are emerging in the agentic workflow. 1.    Reflection 2.    Tool use 3.    Planning 4.    Multi-agent collaboration If you are interested in learning more, instead of me explaining each of these patterns, I would encourage watching the video :) #ai #agent #agenticworkflow #gpt

22

New Video: The role of Generative AI in Integration https://lnkd.in/gQWd5Pgm #LogicAppsAviators

56

1 Comment

Leverage Copilot in Whiteboard to take your team brainstorming to the next level! Copilot will help summarize and even generate new ideas. #msftadvocate #ai #copilot https://lnkd.in/gbRwdKQH

4

1 Comment

Embrace the Future: It’s Time to Consider and Build Your Own AI Agents to Increase Efficiency! 🚀

4

Exciting news! From the White House: "Microsoft will invest $3.3 billion in a new datacenter. Microsoft is committed to creating good-paying jobs with growth pathways and comprehensive benefits such as paid family leave. Microsoft will pair its datacenter investment with a commitment to investing in innovation and workforce in Racine and statewide. Microsoft will partner with Gateway Technical College to develop a Datacenter Academy that trains 1,000 Wisconsinites for datacenter and STEM roles by 2030, and will employ up to 2,000 people in permanent roles at its Racine facility. And, Microsoft will build a Co-Innovation Lab in southeast Wisconsin, as well as team up with the startup accelerator Gener8tor to train 1,000 business leaders to adopt AI in their operations, so that Wisconsin manufacturers can seize AI’s promise. Together, these investments will position the industrial heartland to lead the way in industries of the future."

88

2 Comments

Today is an exciting day, because the cover was lifted on Coplilot+ PCs, and I can finally talk about all the incredible work that my team put into the core platform of these Qualcomm Snapdragon X-based PCs, supporting Copilot and everything else our customers throw at them! Many of our team's innovations are making their debut on Arm with these devices! USB4 support brings 40Gbps of bandwidth, support for multiple 4K monitors, and compatibility with just about any USB-C dock or device you can plug in. Bluetooth Low Energy (LE) Audio delivers lower latency, better battery life, increased voice call quality, and support for LE Audio Hearing Aids. And many of the PCs announced today feature Adaptive Color displays and Presence Sensing experiences powered by on-device AI. These devices do it all - including great performance and battery life! #windows11 #arm #qualcomm #snapdragon #usb4 #bluetooth #leaudio #hearingaids #accessibility #presencesensing #copilot

39

Think of your prompts with Copilot as a conversation. Watch this quick video to learn smart techniques for refining your prompts and guiding Copilot to give you just what you need. #msftadvocate #ai #copilot https://lnkd.in/gm6ViP-p

1

Follow this tutorial to learn how to build a retrieval-augmented generation solution using #OCI #GenerativeAI. https://lnkd.in/ek3kSniK

7

Experience Mixtral 8x22B and Mistral Large API endpoints, powered by NVIDIA NIM microservice, from our NVIDIA API catalog ➡️ https://lnkd.in/gXsNpu2J These performance-optimized #generativeAI models by Mistral AI, excel in complex multilingual reasoning tasks, and more.

4

Do more in Microsoft Mesh with data and AI

1

🚀 𝗘𝘅𝗰𝗶𝘁𝗶𝗻𝗴 𝗡𝗲𝘄𝘀 𝗳𝗿𝗼𝗺 𝗠𝗶𝗰𝗿𝗼𝘀𝗼𝗳𝘁 𝗔𝘇𝘂𝗿𝗲 𝗔𝗜 𝗠𝗼𝗱𝗲𝗹 𝗖𝗮𝘁𝗮𝗹𝗼𝗴! 🌐 In partnership with Meta, Microsoft is thrilled to unveil the next-generation Meta Llama 3 models on Azure AI Model Catalog! 𝗜𝗻𝘁𝗿𝗼𝗱𝘂𝗰𝗶𝗻𝗴 𝗠𝗲𝘁𝗮-𝗟𝗹𝗮𝗺𝗮-𝟯-𝟴𝗕-𝗜𝗻𝘀𝘁𝗿𝘂𝗰𝘁 𝗮𝗻𝗱 𝗠𝗲𝘁𝗮-𝗟𝗹𝗮𝗺𝗮-𝟯-𝟳𝟬𝗕-𝗜𝗻𝘀𝘁𝗿𝘂𝗰𝘁 - your gateways to unparalleled AI performance and innovation. 🧠 These state-of-the-art models are pre-trained and fine-tuned with instructions to deliver significant enhancements in AI capabilities. Thanks to extensive training on a massive dataset and advanced parallelization techniques, Meta Llama 3 models promise to revolutionize how developers, enterprises, and digital natives leverage AI, offering improvements in reasoning, language understanding, and much more. 💡 Explore the endless possibilities with Meta Llama 3 in the Azure AI Model Catalog. Whether for text summarization, content creation, or conversational AI, these models are designed to scale and adapt to a variety of use cases. 🌟 𝗚𝗲𝘁 𝗦𝘁𝗮𝗿𝘁𝗲𝗱: Dive into the future of AI by accessing these models today on Azure AI. Explore, deploy, and integrate seamlessly using Azure's comprehensive tools and infrastructures: 1) 𝗘𝘅𝗽𝗹𝗼𝗿𝗲 𝗠𝗼𝗱𝗲𝗹𝘀 𝗶𝗻 𝗔𝘇𝘂𝗿𝗲 𝗔𝗜 𝗦𝘁𝘂𝗱𝗶𝗼: https://lnkd.in/dmrQhWZi 2) 𝗙𝗼𝗿 𝗺𝗼𝗿𝗲 𝗱𝗲𝘁𝗮𝗶𝗹𝘀, 𝗿𝗲𝗮𝗱 𝘁𝗵𝗲 𝗳𝘂𝗹𝗹 𝗯𝗹𝗼𝗴 𝗽𝗼𝘀𝘁: https://lnkd.in/ddrq9Emw #AI #AzureAI #MetaLlama3 #Innovation #MachineLearning #CloudComputing #Microsoft #productmanagement

8

1 Comment

Many of us at Microsoft have been lucky enough to have been using Copilot for a year, along with hundreds of early customers, and we've learned a ton along the way. Now, we're excited to share what we've learned about AI @ Work as our own "customer zero". See below for some quick hits on how we did it, but don't miss the full read below to learn how you can do it too! #Copilot #Microsoft365 #MSFTAdvocate ✔ Go for the big wins (and the easy ones too) ✔ Find your internal champions ✔ Double down on skilling up ✔ Build the AI habit

3

Small models becoming smarter! Phi-3 reinforces the importance of data quality.

24

🌟 Leading Innovation with GenAI at Amazon! 🌟 As the Product Manager for our AI-enhanced catalogue, I'm thrilled to share how we're using generative AI and A/B testing to revolutionize online shopping. Our advanced analytical techniques, including causal random forests and Bayesian structural time series, are key to evaluating the impact of GenAI. Explore our innovative approaches in this detailed Amazon Science article. #GenAI #ABTesting #Ecommerce #Innovation #AmazonScience

27

1 Comment

When greentech unicorn Enpal needed to automate its solar panels sales process, it turned to custom AI software specialists dida Machine Learning. Taking an abstract mathematical approach to complex ML models, dida used Google Cloud to build an ML solution that cut Enpal’s process time by 87%, saving the company time and reducing costs while improving its customer experience. Learn more on the blog ⬇️

3

1 Comment

Huge milestone: Microsoft released its inaugural report on #AI transparency this week. As a #ResponsibleAI champ, I’m proud to be part of our commitment to putting people first in our innovations. This report highlights practices & processes #Microsoft uses to create a future where AI benefits everyone.

18