Jeremy Y. Chen

Congrats the team on ICML acceptance of the Wukong paper: https://lnkd.in/g-MyahNH, which represents a breakthrough of recommendation model design and efficiency. Recommendation models traditionally suffer from not able to utilize the full computational power of GPU (only 5-10% of peak GPU FLOPS while LLMs can reach >50%), due to their sparsity. Wukong is first of its kind by pushing toward 40% GPU peak FLOPS though model-infra codesign. One of my teams consisting of GPU experts is building a high performance model component libary where we’ll provide highly optimized model components (eg sequence summarization, x-sequence interaction, x-GPU layout scheme) that could be composed - with the goal that models built from this library have high performance out of box to scale the model-infra codesign effort. We are also scaling training beyond the limits that are typical of recommendation models - from hundreds of GPUs to 10s of thousands. If this is the kind of work that excites you, come join us!

264

4 Comments

Calling all IT leaders, developers, and practitioners! Ready to explore the future of data in the age of GenAI? Join Google Cloud for Database Day: Sunnyvale on August 27th! **🔥 Here's what you can expect: ** Two tracks packed with insights: one for IT Leaders & Executives, and another for Developers & Database Practitioners. Expert-led sessions on topics like building GenAI apps with Google Cloud databases, maximizing PostgreSQL ROI, and mastering Cloud SQL. Hands-on labs and demos to level up your skills and connect with peers. Don't miss this chance to build your GenAI future. Register now - spots are filling up fast! #GoogleCloud #DatabaseDays #GenAI #DataInnovation

1

ZGC: Generational Mode is proposed to be the default ZGC mode for JDK 23 :-) Highly recommending you try ZGC!       https://lnkd.in/gAJC6GYY #zgc #java #openJDK #GC

34

1 Comment

Thrilled to support 💪 Setting Customer TPM/RPM Limits on Admin UI - https://lnkd.in/gCVW3C7x (+4 more updates 👇) ⚡️ UI - See Time to first token per model s/o Lucas M. ✨ CRUD endpoints for customer budgets 🛠️ Slack - Fix slack daily deployment reports s/o Marek Keřka 🛠️ UI - Fix model latency + hanging request metrics 

1

Great compilation of resources! These are definitely some of the best materials out there for prepping for tech interviews. I've personally found the "System Design Interview" books to be incredibly helpful for grasping complex design principles. Also, "Neetcode" has a great collection of problem solutions that are super handy for quick review sessions. One resource I'd add to the list is "The Pragmatic Programmer" book. It's not directly interview-focused, but it provides invaluable insights into software development that can help deepen your understanding and improve your problem-solving skills. For behavioral interviews, practicing the STAR method with real-life examples from your experience can be a game-changer. It's all about storytelling and demonstrating your impact. Curious to hear what others think about these resources. What has been the most effective material for your prep? #TechInterviews #InterviewPrep #SoftwareEngineering

2

Thats the exact explanation I gave to a group of investors last week. (… it’s just Precision/Recall, attributes vectorization as “embedding”, K-Nearest Neighbor, User Search Intent, DAG, ML algorithms, etc all over again …). LOL and they asked what’s new this time around. And I was like not much, ML is still the backbone, and AI is the inference and actions based on the model. It works better when there is a feedback loop in a closed system. But the current state of the art like ChatGPT is based on scraping the Internet (garbage in, garbage out). Kinda like early search engines. What is a better analogy is that NVIDIA/Snowflake etc are selling tools, aka pickaxes to prospectors. The gold is up to you to find. And the opportunity right now is to monetize industry specific (healthcare, travel, financial services) clean data to build models with (aka a treasure map for the prospectors). From an enterprise perspective, what’s out there today aka generative AI isn’t very useful. Too much hype. To build useful engines will require an effort similar to search revolution 2 decades ago. From

16

3 Comments

LLMs and LangChain: The Dynamic Duo 🦸‍♀️🤖 Just finished a fun project where we built a Q&A bot using these two powerhouses. It's like having a personal AI assistant that's always ready to answer your burning questions! 🔥 Here's what I learned: LLMs are wizards: They can conjure up answers from thin air (or, uh, the internet). 🧙‍♂️ LangChain is the glue: It holds everything together and makes it easy to use. 🔗 Endless possibilities: This tech can be used for anything from customer support to writing catchy social media posts. 📝 Want to see it live (obvio a test server link 😛 ): https://lnkd.in/dZNAVCSV Note: it's a render link, they will take down the instance because of no use, you can hit the link and wait for two minutes, they will spin up the instance and you can try it out. Want to see the code? Check it out here: https://lnkd.in/dzjh4r47 #AI #MachineLearning #NLP #LangChain #LLM #Python #Developer #FunProject"

22

Lunar.dev is for Lambdas too. API calls in AWS Lambda can be a traffic jam, impacting reliability, speed, and cost. Discover how Lunar.dev smooths the ride by both providing visibility and offering proactive remediations for external API errors. 🔗 Read the full blog for a smoother API Consumption journey!

30

To smooth your reading experience, let me give a bit of background; I am planning to write tech posts here (irregularly), to share the AI/ML/tech exp between all of us. I came from Hong Kong, where I wrote more on political / cultural, or even dining / winery matters, because that was our major concern, and I would not receive any feedback on any of my tech posts anyway.... but here in North America things are different. I joined some IT Whatsapp groups and the suggestions to problems had been really inspiring. When the democracy foundation is strong, and when the basic ethics are there, we can focus on the state-of-the-art tech, which hopefully can change our lives for the better. So in the next or coming posts, I would formulate some objectives or problems, which I may have a suggested approach solution, but I would like some of you to share your thoughts. It's just like beer talk. I am expecting your words in my coming posts!!

12

2 Comments

I am excited to share the latest blog, co-authored by myself, Junhwi Kim and Vyas Sekar, detailing how we managed to successfully localized and improved LLM by combining Fine-Tuning and RAG. https://lnkd.in/gKDkgGPq

70

3 Comments

🌟What’s your #AI superpower?🌟 I took the @slackhq AI persona quiz and discovered that I’m the AI Superfan. Take the quiz to find out what member of the #AITeam you are and how you contribute to the #FutureOfWork 🚀 https://lnkd.in/gqkwcAHy

1

Databricks Clean Rooms, powered by Delta Sharing, enables organizations across industries to collaborate with partners in a privacy-safe manner without relinquishing control over sensitive data. Key features: ✔️ Any language, any workload ✔️ Any cloud, no replication ✔️ Any scale, any trust level Learn how it works👇

18

Hey there, Domo users quick update coming your way!  We're super excited to announce this: you can now harness external AI models in Magic ETL. Our new AI Inference tile lets you deploy any of your eligible externally hosted models, like Databricks and Amazon Bedrock, within Magic ETL using Domo’s ML connectors.  (Note that the models will need CSV input and output formats. We plan to expand in the near future to include JSON-based models in the near future.) I'm stoked to share this update, being able to supercharge your ETL with your own model will give you the ability to elevate your data game like never before!  Curious about what this means for you? Dive into all the details right here: https://lnkd.in/gcguUqs6 #domoupdates #domoproduct 

20

1 Comment

Miss the conversation between Mark Zuckerberg and Jensen Huang at SIGGRAPH? Watch the whole conversation as they dive into the latest advancements in generative AI and open-source software, and explore how these technologies are empowering developers and creators to build new and innovative experiences. #GAI #TheNextComputingPlatforms #opensource #opencompute

2

🚀 We've open-sourced TimesFM – a decoder-only foundation model for time-series forecasting 📈 Key Highlights: ✅ Large-Scale Training Data: Trained on a corpus of over 100 billion real-world time points, including Google Trends and Wikipedia pageviews. ✅ Zero-Shot Forecasting: Offers high-quality forecasts even on unseen datasets, without requiring fine-tuning or additional training. ✅ Outstanding Benchmark Performance: Achieved competitive results across a comprehensive suite of benchmarks. 🤗 Explore the model on Hugging Face and GitHub. I'm encouraged by the early feedback and look forward to seeing how #TimesFM will be used in real-world applications for forecasting and anomaly detection. 🙏 A special thanks to the talented research team behind this achievement: Abhimanyu Das Rajat Sen Yichen Zhou Weihao Kong

44

1 Comment

🚀 Insights from Pathways to Tech Event #16 🚀 I recently had the privilege of attending the Pathways to Tech event, featuring an incredible lineup of industry leaders: Dennis Crowley - Founder of Foursquare ($150M Revenue) Tariq Afaq - First Engineer at AppLovin ($30Bn IPO) Julian Alvarez - Former SWE at Meta, CEO of Wisdolia ($20K+ MRR) Hamza Ali - CEO of Olostep (Hiring Headstarter Fellows) Erik - ML Engineer Intern at TRC, CS Creator with 75K followers Here are some of my key takeaways: 🔹 Dennis Crowley emphasized the power of team collaboration and the importance of creating emotionally connective products. His reflections on the role of software engineers in the AI age and the challenges of building a product with a potential exit in mind were particularly enlightening. 🔹 Julian Alvarez shared his journey from coding as a kid to founding Wisdolia. His advice on minimizing startup mistakes and focusing on specific user groups to drive business growth was truly inspiring. 🔹 Hamza Ali discussed the challenges of fine-tuning AI models and his mission at Olostep. His insights into using AI to enhance user experience were thought-provoking. 🔹 Tariq Afaq and Erik provided valuable perspectives on scaling tech companies and the role of personal motivation and growth in the tech industry. Favorite Speaker: Dennis Crowley’s discussion on the human elements of startups and impostor syndrome was particularly impactful. His focus on maintaining good health and staying updated with new tools resonated deeply with me. If you missed the event, you can catch the full video here: https://lnkd.in/d5-ZYukg Huge thanks to @ Headstarter for organizing this insightful event! 🙌 #TechEvent #PathwaysToTech #Foursquare #AppLovin #Wisdolia #Olostep #AI #Startups #TechIndustry #Headstarter

2

The Power of AI-Driven Pair Programming As a Staff Engineer at Artium, I've witnessed firsthand the transformative power of integrating AI tools, particularly Large Language Models (LLMs), into pair programming. This innovative approach is revolutionizing the way we code and collaborate, and I'm excited to share my insights with you. One of the most significant benefits of utilizing LLMs like GPT in pair programming is the dramatic acceleration of coding processes. By providing real-time feedback and code suggestions, these AI models reduce the time spent on routine tasks and debugging, allowing teams to push updates and features at an unprecedented pace. This increased speed and efficiency is a game-changer in the fast-paced world of software development. Moreover, LLMs leverage vast datasets to suggest coding solutions that adhere to best practices, significantly improving the quality and reliability of software products. This AI-driven guidance helps minimize errors and enhance overall code performance, ensuring that we deliver top-notch solutions to our clients. Another crucial aspect of AI-driven pair programming is the optimal utilization of resources. By handling repetitive tasks, AI allows developers to focus on complex problem-solving and creative innovation. This shift not only optimizes human staff allocation but also amplifies the creative potential of development teams, leading to more groundbreaking solutions. AI models in pair programming also serve as an on-demand knowledge base, teaching new frameworks and languages, and keeping the team updated with the latest industry standards. This fosters a culture of continuous learning and professional growth, which is essential in the ever-evolving tech landscape. Furthermore, with AI, the benefits of pair programming are no longer confined to office hours. Developers can interact with AI partners anytime, ensuring continuous productivity and making software development more adaptable to varying work schedules. This flexibility is crucial in today's diverse and dynamic work environments. In my experience, the AI-human partnership in software development not only increases operational efficiency but also paves the way for more innovative solutions. As we continue to explore and understand AI's potential, it's clear that the fusion of AI with human ingenuity in pair programming is just the beginning of a larger transformative journey in tech. I'm curious to hear your thoughts and experiences. What AI tools have you integrated into your development process, and how have they transformed your workflow? Let's discuss how we can further harness AI to enhance our productivity and creativity in tech. #SoftwareDevelopment #PairProgramming #AIinTech #Innovation #TechCommunity

5

Istio Service Mesh: great to see Ambient moving up to Beta, delta xDS becoming the default, and api-gateway becoming a first class citizen in Istio. https://lnkd.in/gYPmnTuj

34

The afternoon Keynote @ Stripe Sessions was Fireside Chat between Patrick Collison (CEO of Stripe) and Jensen Huang (CEO of NVIDIA). Patrick was an excellent moderator — asked Jensen really probing, nuanced questions that Jensen answered with his usual sharp wit and unique (and impressively well rationalized) perspective. A couple of quotes among many that stuck with me: - “Current CPU installed market is $1 trillion of hardware, all of which will be upgraded to GPUs over the next 5 years. And I would estimate about another $1 trillion will be spent on top of that during the same time period as companies try to do more with AI. Add those together — and I think at least $2 trillion will be spent on GPUs in the next 5 years.” - “I like to go after Zero-Billion Dollar Markets — nonexistent markets that I envision & can rationalize well, but will be very hard, expensive & time-consuming to achieve.”

1