- Видео 489
- Просмотров 552 075
Alex Chao
Добавлен 5 сен 2013
Tech, AI, and Life from your resident Chaos coordinator.
Currently, I’m building next-generation AI products at Microsoft through the Office of the CTO with an emphasis on multimodal and large language models.
Over the last decade, I’ve successfully launched applied ML products across industries in health, gaming, finance, defense, energy, autonomous vehicles, semiconductors, and enterprise tech.
I’ve held product management roles at SambaNova Systems and C3.ai building out ML Platforms and deep learning offerings to enterprise customers. I continue to be an active ML practitioner and was the first data science hire at Uber’s Advanced Technologies Group working on building large scale computer vision and sensor fusion algorithms for self-driving cars.
Proud alum of UC Berkeley! Go Bears!
Currently, I’m building next-generation AI products at Microsoft through the Office of the CTO with an emphasis on multimodal and large language models.
Over the last decade, I’ve successfully launched applied ML products across industries in health, gaming, finance, defense, energy, autonomous vehicles, semiconductors, and enterprise tech.
I’ve held product management roles at SambaNova Systems and C3.ai building out ML Platforms and deep learning offerings to enterprise customers. I continue to be an active ML practitioner and was the first data science hire at Uber’s Advanced Technologies Group working on building large scale computer vision and sensor fusion algorithms for self-driving cars.
Proud alum of UC Berkeley! Go Bears!
How do we build a future of AI agents? | AI Conference (2024)
Watch the panel I moderated at the AI Conference 2024 on Building AI Agents with fellow speakers Anna Marie Clifton from Zapier and Shelby Heinecke of Salesforce!
Discover the key principles and methodologies in designing and implementing intelligent agents as we share our experiences in developing various types of agents and discuss the challenges and solutions to building robust, production-ready agentic systems.
Leave a comment below on how you are building AI agents!
And follow me on other platforms so you’ll never miss out on my updates!
💌 Sign up for my free AI newsletter Chaos Theory: alexchao.substack.com/subscribe
🐦 Follow me on Twitter alexchaomander
📷 And Instagram! insta...
Discover the key principles and methodologies in designing and implementing intelligent agents as we share our experiences in developing various types of agents and discuss the challenges and solutions to building robust, production-ready agentic systems.
Leave a comment below on how you are building AI agents!
And follow me on other platforms so you’ll never miss out on my updates!
💌 Sign up for my free AI newsletter Chaos Theory: alexchao.substack.com/subscribe
🐦 Follow me on Twitter alexchaomander
📷 And Instagram! insta...
Просмотров: 448
Видео
GraphRAG Open-Source Release Announcement
Просмотров 7 тыс.4 месяца назад
The team at Microsoft is pleased to announce that GraphRAG is now available in open-source! GraphRAG is a research project from Microsoft exploring the use of knowledge graphs and large language models for enhanced retrieval augmented generation. It is an end-to-end system for richly understanding text-heavy datasets by combining text extraction, network analysis, LLM prompting, and summarizati...
UFO: A UI-Focused Agent for Windows OS Interaction
Просмотров 1,4 тыс.5 месяцев назад
Watch my colleague Chaoyun Zhang present his team's research and open-source project UFO: A UI-Focused Agent for Windows OS Interaction. UFO is a UI-Focused dual-agent framework to fulfill user requests on Windows OS by seamlessly navigating and operating within individual or spanning multiple applications. Check out the repo here: github.com/microsoft/UFO/ And the paper! arxiv.org/abs/2402.079...
Bridging the Gap Between AI Research and Production with Semantic Kernel
Просмотров 3 тыс.5 месяцев назад
AI experts and enterprise development teams often use different tech stacks with different concepts and resource types. Now that Semantic Kernel is V1.0 in Python, C#, and Java though, your teams now have the opportunity to speak the same language. With Semantic Kernel, ML teams can create the same assets for plugins, planners, and agents, evaluate them, and then give them to app dev teams to d...
Guidance: Towards Controlling the Output of Large Language Models
Просмотров 8306 месяцев назад
Watch my colleague Harsha Nori present on the open-source research project Guidance! Originating at Microsoft, Guidance is a programming paradigm that offers superior control and efficiency compared to conventional prompting and chaining of large language models. It allows users to constrain generation (e.g. with regex) as well as to interleave control (conditional, loops) and generation seamle...
ML Agent Bench: Evaluating Language Agents on ML Experimentation (Paper Review)
Просмотров 4156 месяцев назад
Check out the ML Agent Bench from Stanford as a way to benchmark AI agents on experimentation tasks! github.com/snap-stanford/MLAgentBench This is part of an ongoing series called the Pals of Autonomous Agents where I engage with the AI community to talk about how we are building the future of autonomous agents. Make sure to subscribe and share your thoughts in the comments below! And follow me...
GraphRAG: LLM-Derived Knowledge Graphs for RAG
Просмотров 127 тыс.6 месяцев назад
Watch my colleague Jonathan Larson present on GraphRAG! GraphRAG is a research project from Microsoft exploring the use of knowledge graphs and large language models for enhanced retrieval augmented generation. It is an end-to-end system for richly understanding text-heavy datasets by combining text extraction, network analysis, LLM prompting, and summarization. For more details on GraphRAG che...
(Demo) Using GraphRAG on Microsoft CTO Kevin Scott podcasts
Просмотров 1,6 тыс.6 месяцев назад
Watch this demo of how we can use LLM-created knowledge graphs applied to a dataset of transcripts from Kevin Scott's "Behind the Tech" podcasts to answer questions that traditional RAG systems would fail at! GraphRAG is a research project from Microsoft exploring the use of knowledge graphs and large language models for enhanced retrieval augmented generation. It is an end-to-end system for ri...
Artificial Intelligence Controller Interface (AICI) Explained!
Просмотров 7677 месяцев назад
Artificial Intelligence Controller Interface (AICI) Explained!
What can AI agents learn from past efforts at automation?
Просмотров 2328 месяцев назад
What can AI agents learn from past efforts at automation?
03.06.2024 - Semantic Kernel Office Hours (US/Europe Region)
Просмотров 6858 месяцев назад
03.06.2024 - Semantic Kernel Office Hours (US/Europe Region)
02.22.2024 - Semantic Kernel Office Hours (US/Europe Region)
Просмотров 3889 месяцев назад
02.22.2024 - Semantic Kernel Office Hours (US/Europe Region)
TaskWeaver: An AI Agent Framework for Data Analysis
Просмотров 9099 месяцев назад
TaskWeaver: An AI Agent Framework for Data Analysis
02.21.2024 - Semantic Kernel Office Hours (US/Europe Region)
Просмотров 3489 месяцев назад
02.21.2024 - Semantic Kernel Office Hours (US/Europe Region)
Driving Hyper Engagement using Generative AI - Stepwise Planner v2 | Semantic Kernel Community
Просмотров 4009 месяцев назад
Driving Hyper Engagement using Generative AI - Stepwise Planner v2 | Semantic Kernel Community
SK in a Box | Semantic Kernel Community
Просмотров 4059 месяцев назад
SK in a Box | Semantic Kernel Community
Azure Chat Solution Accelerator Demo | Semantic Kernel Community
Просмотров 2,1 тыс.9 месяцев назад
Azure Chat Solution Accelerator Demo | Semantic Kernel Community
Jeremy Toeman: Empowering Creators To Make Compelling Videos with AI | Humans of AI #14
Просмотров 2289 месяцев назад
Jeremy Toeman: Empowering Creators To Make Compelling Videos with AI | Humans of AI #14
02.15.2024 - Semantic Kernel Office Hours (Asia-Pacific Region)
Просмотров 2399 месяцев назад
02.15.2024 - Semantic Kernel Office Hours (Asia-Pacific Region)
02.14.2024 - Semantic Kernel Office Hours (US/Europe Region)
Просмотров 4839 месяцев назад
02.14.2024 - Semantic Kernel Office Hours (US/Europe Region)
Semantic Kernel 2024 Spring Roadmap!
Просмотров 1 тыс.9 месяцев назад
Semantic Kernel 2024 Spring Roadmap!
02.07.2024 - Semantic Kernel Office Hours (US/Europe Region)
Просмотров 3329 месяцев назад
02.07.2024 - Semantic Kernel Office Hours (US/Europe Region)
01.31.2024 - Semantic Kernel Office Hours (Asia-Pacific Region)
Просмотров 1249 месяцев назад
01.31.2024 - Semantic Kernel Office Hours (Asia-Pacific Region)
01.31.2024 - Semantic Kernel Office Hours (US/Europe Region)
Просмотров 3079 месяцев назад
01.31.2024 - Semantic Kernel Office Hours (US/Europe Region)
01.24.2024 - Semantic Kernel Office Hours (US/Europe Region)
Просмотров 34210 месяцев назад
01.24.2024 - Semantic Kernel Office Hours (US/Europe Region)
01.17.2024 - Semantic Kernel Office Hours (Asia-Pacific Region)
Просмотров 26910 месяцев назад
01.17.2024 - Semantic Kernel Office Hours (Asia-Pacific Region)
01.17.2024 - Semantic Kernel Office Hours (US/Europe Region)
Просмотров 31710 месяцев назад
01.17.2024 - Semantic Kernel Office Hours (US/Europe Region)
01.10.2024 - Semantic Kernel Office Hours (US/Europe Region)
Просмотров 74810 месяцев назад
01.10.2024 - Semantic Kernel Office Hours (US/Europe Region)
Semantic Kernel, Langchain, Promptflow | Where does SK fit in the AI ecosystem?
Просмотров 6 тыс.11 месяцев назад
Semantic Kernel, Langchain, Promptflow | Where does SK fit in the AI ecosystem?
Thank for sharing👍
it's so bad that it continues to interrupt you when you use it. It's local and yet wants to ensure your safety in al ways, continually policing you.
animals don't have language
hi alex, i am currently doing a stage and your video helped me thanks:), but i had a question, i dont understand what is the best method to obtain all the embedding, and i dont seem to find the api to know what should i put in the body of the request , can you help me?
I have been using langchain for a long time, and I honestly dont see how this differs to it . I understand there is a slight implementation difference. but apart from that only other benefit I see is C# support
You could save yourselves the political/propaganda element. From all uncounted number of articles available, the choice of this particular topic is more than flashy. It's ridiculous for techie persons who are expected to be smart in general...
Why are Elon, Democrats, and Republicans authorship hard coded into the algorithm? How is that "the town square" ?
If any call is questioned the electronic tracking system can check.
All it is, they are eliminating people they have to pay ! That is ALL this is about !
Hi @alexchaomander, Is there a python version instead of .NET version as a backend, I have being looking for it, and could not find. Can you point me to any github or knowledge base?
Any upadte on that streamlit code , that would be helpful thanks
I started watching this a year ago when you dropped this series and I am coming back to it to share with other developers and such. Since then we implemented some awesome features in our own software and creating a weird collaborative environment that is using semantic Kernal to edit the documents generated by multiple agents and people in a single environment. Its bee a wild journey and we found some unexpected features so just wanted to say thank you for these videos!
Simple, clear, and useful. Thanks a lot
Love that SK supports C# in addition to Python and JAVA. It is a really clean architecture that merges so well in existing .NET Core apps.
Hi, how about the abstract ideas visualization in multidimensional space eg. some mathematical ideas and relations to other ideas - eg. in form of knowledge graph and/or deeper details research possibilities tools? Are there any such tools based on ai?
great research topics but as an hands-on nlp engineer on ner boosted knowledge graphs and LLMs , my experiences say too naive to believe that it would work in productions systems
Great Demo!
I already know Apples mason rat and its escape route to Australia. Why does every name i look up its always some software engineer or IoT tech... AI engineer... it doesnt stay believable. I've been waiting for Tara - a rat(arat) to be useful. I guess it did.
Streamlit code would be great, thanks
It's crap. No where near as nuanced for modern war as necessary.
Very nice presentation and explanation
Are there any accelerators to convert a typical knowledge corpus of unstructured text to a knowledge graph conducive for graph rag? I understand we need to extract entities and figure out relationships but who does that work? An LLM?
I think this is a similar situation when Kodak was trying to sell film in a digital world. When they tried to pivot to digital, it was too late. Chegg is trying to pivot to AI, but it might be too late with too much competition. The value Chegg offered 5 years ago is not the same value today.
I dont like chris Christie but that was actually pretty funny. Vivek got him even better on the return
Is there a demo of how to setup it?
Very buggy. Even fundamental parameters as temperature, n, top_p are ignored !!! Expects next release.
Is this still an issue?
Thanks for the video. I can see a Use Case in my energy industry. Does GraphRAG work across all "modes" and "modalities"?
first of all, there is no MOVEMENT but state sponsored Russian proxies like the Yemenis. a very wrong choice of a dataset. second thing - there is NO NOVOROSSIA
This is agreat demo Alex. I am going to try it out today
Wow, I feel like I got a memory lapse, I am exactly the same person I was 42 seconds ago
Thank you for your info.
This is amazing!
Its a month later, where's the code you promised? Please?
Finally 🎉
Admit it: y'all built graphrag first for use by the CIA. This is not a joke.
Where did you get the body for this? This whole text is taken from some chief Russian propaganda bureau 😅
this video could have been a tweet
I call dip!!!!
appreciate the open sourcing! keep it up :)
This is powerful!
Pretty soon, everyone will be graphragging their podcasts. Jre will be neat.
Why do we need other faces on screen ? Hope they know they are just distractions :)
And I hope you know that you can zoom in to the screen to not see them and that it's always better not to say anything if you don't have anything nice or useful to say 😊
è bellissimo!
Excellent! Can't wait to give this a try.
put it in kernelmemory...
Python 😞
Amazing, I’ve been stoked about this! Thanks! 👍🏻
To learn more about GraphRAG, watch the deep-dive on the tech behind the system with the team here! ruclips.net/video/r09tJfON6kE/видео.html
But how to do it? Lots of videos are there, but none of them telling actually how to do it?
Just came across this, right after my break. Thanks for sharing it :)