Eric Charles

Your Contribution for tonight ✨: Airbyte ⬇️ Co-founded by Michel Tricot and Jean (John) Lafleur, Airbyte is a data integration platform for ETL / ELT data pipelines. You can integrate from APIs, databases, data warehouses, data lakes and data lakehouses. Airbyte is available self-hosted and cloud-hosted. 🧪 Here’s why you should contribute to this repo: 1️⃣ Their tech leverages modern practices to help solve data engineering challenges, affecting hundreds of companies and data professionals. If you are a Python or Java developer and want to further contribute to a powerhouse repo (with over 13,000 stars!), this is your opportunity. 🦾 2️⃣ The repo "really" welcomes new contributors. Many projects have a contributing.md file but rarely accept new contributors. Airbyte brought on their repo 12 new contributors just this last month! If you are looking for a welcoming repo, Airbyte may be what you have been looking for. 😄 3️⃣ Nearly ~45% of contributors submit a second PR. This means the repo has a supportive environment for building upon past contributions. If you enjoy working on this repo, chances are you can continue to work on it! ⚡️ If you are interested in contributing to building the only tool devs need to move data, check more of Airbyte's stats on Quira (https://lnkd.in/eM7etquW) or access the GitHub page directly here: https://lnkd.in/e_4Gczrk

13

⭐️ New video release 📺: Performant, scientific computation in 🐍Python and 🦀 Rust Watch Stefan Ulbrich as he guides you through building performant scientific computation packages in Python and Rust, offering insights into efficient computation and optimization for machine learning and data science applications. 📺 Watch the video on YouTube: https://lnkd.in/eq-BhHdH Stefan Ulbrich, Senior Data Scientist, offers a tutorial on enhancing scientific computation in Python and Rust. The session explores packaging numerical calculus and algorithms using modern tools. Highlighting the efficiency of vectorization in Python for computational tasks within Machine Learning, Ulbrich presents the integration of Rust to optimize package performance. Rust, recognized for its performance and memory safety, complements Python, particularly for instances prioritizing speed and modern language features. The talk demonstrates creating a scientific Python package, gradually transitioning components to Rust for enhanced performance. Through an example of implementing the "Expectation maximization for Gaussian Mixture Models" algorithm, attendees learn how to leverage modern tools to build, optimize, and extend packages efficiently. Ulbrich emphasizes that while Rust knowledge is not mandatory, the session provides a high-level understanding of its benefits for Python extension and performance improvement. For a hands-on experience, participants are encouraged to clone Ulbrich's repository for swift installation and to optimize session time effectively.

9

1 Comment

Happy to represent Segments.ai (YC W21) at today’s Flanders Make’s Symposium! Sam Dehaeck & Xavier Jordens presented a demo setup to showcase the FARAD2SORT (Fast Deep Learning and Deployment for Products Sorting) research project. TicTac’s were counted and sorted based on color. Unknown boxes were separated, and automatically uploaded to Segments.ai for labelling before being pulled back into re-training. Garben Tanghe also represented VINTECC as a partner in the project to accelerate the used ML models on the edge device. #MachineLearning #DeepLearning #AI #Manufacturing

59

3 Comments

Looking forward to the insights. ---------------------- 💥 I am Jeroen. 💥 Supporting your connected data experiences 💥 Providing training, consultancy and keynotes Liked this post? Want to see more? 🔔 Ring it on my Profile Follow #connecteddata 🔝 Connect with me #solita #api #apiops #sharingiscaring

7

Tomorrow I will be at the AI unboxed event of Agoria and Sirris | Innovation forward. A complete afternoon and evening full of insights on AI and how to embrace it within you company. Join the experts and other companies to learn more on how they are integrating AI within a digital transformation journeey. Come say hi 👋 at ourour booth! #codit #ai #azureai #agoria #sirris #mircosoft #azureai #digitaltransformation #azure

9

1 Comment

The first blog about #mlops from Zero to Hero is now available on both #medium & thaink² blogs thaink² blog: https://lnkd.in/eXKynJhy medium : https://lnkd.in/e9are6gp Your feedbacks are welcome 😜 #ml #mlops #datascience #rstat #tidymodels #vetiver #plumberapi

18

1 Comment

Today is the first OpenAI DevDay for this autumn. No new models are expected, but OpenAI has promised to introduce some long waited features to APIs. This is the schedule (in San Francisco timezone): - 10:00 - Welcome + kickoff by Olivier Godement & Romain Huet - 11:00-11:45 - Topic to be revealed during keynote - 12:00-12:45 - Topic to be revealed during keynote - 12:45-12:00 - Lunch - 2:00-2:45 - Topic to be revealed during keynote - 3:00-3:45 - Topic to be revealed during keynote - 4:00-4:45 - Fireside chat with OpenAI leaders - 4:45-7:00 - Closing reception So reading this, there could be up to four new announcements coming. Here is my wishlist in priority order: - Audio/voice multimodal API for gpt-4o (so advanced voice mode API) - Function call webhooks, so that AI could call app function in mids of voice-call or o1 reasoning sessions. - New gpt-4o image generation API - Prompt caching - Web search enabled api (ala Perplexity or Command-r) But I hope they leave the best for London in end of month, where I'm going 😂

12

1 Comment

The Quantum era isn’t a distant dream, it’s happening right now, and it will redefine how industries operate. 👀 These real-life applications are already reshaping how businesses operate, pushing industries toward unprecedented levels of efficiency and innovation. 🔎 Don’t want to get left behind? Get involved with the Quantum community that’s positioning Belgium at the heart of the Quantum innovation map and gain the ability to enter a valuable network of forward thinkers. 🔗 Connect with us: https://lnkd.in/eneE64DV #QuantumApplications #Quantum #Quantum4Belgium ❤️ Quantum Circle

20

1 Comment

📢Github Insights 📈 advanced so far with traffic tool ! I become surprised of a huge instant interest on personal projects (instant splash of ~30 git pulls 🔥) ... even for such that were for a long time ago in a constant maintenance ‼ How to watch your stats 👀 : ✅ Simply navigate to your project page 👉 Insights 👉 Traffic Enjoy your sunday by exploring your project insights! ☕️ PS: With the AREkit in particular (figure below), thanks everyone for interest, this support motivates me and means alot 🙌 Meanwhile, I am head to back on track with the maintenance and updates on AREkit 🚀 AREkit 🌟: https://lnkd.in/g6jyWWqb #github #stat #traffic

1

🚀 Calling all Data Scientists, Data / ML / MLOps Engineers, and everyone interested in data! AE and NannyML have joined forces for an interesting meet-up, and I'm thrilled to invite you! Learn from industry experts as they share their insights about developing machine learning models or data products and using them successfully in production. 📣 Unlocking the Value of Data through Machine Learning. Senne, Data Scientist at AE, will delve into two real-world case studies, demonstrating how machine learning is transforming low-margin businesses. Discover how AE harnesses the power of models like Meta's Prophet and XGBoost to drive impressive business results. 📣 Addressing the Engineering Challenges of Building Data-Centric Products. Join Niels, Lead Engineer at NannyML, as we dissect the engineering hurdles of data-centric Product Development, from inception to deployment. ✏ Registration: https://lu.ma/pdds-leuven 📅 Save the Date: Tuesday, 14th of May 📍 Location: AE, Leuven 🕡 Time: doors at 18:30, talks at 19:00, networking afterwards #meetup #mlops #machinelearning #datascience #leuven

34

13 Comments

Our book "Generative AI for Effective Software Development" is now available on Springer: https://lnkd.in/et6MYb7c You can also pre-order the paper version on Amazon: https://a.co/d/9Xczbtq 🔎 About the Book: an in-depth empirical exploration of how Generative AI is adopted through detailed chapters featuring case studies, surveys, and experiments. Our book provides both practitioners and academics with a nuanced understanding of the current and future impacts of AI technologies in software engineering. 📘 Why Read This Book? Whether you're a developer looking to integrate AI into your projects, a manager aiming to navigate the evolving tech landscape, or a researcher seeking the latest in empirical AI studies, this book is an invaluable resource. 🌐 Get Your Copy Today! Visit Springer's website to purchase your digital or print copy and stay ahead in the rapidly evolving field of software development.

165

Today I had a talk on the main stage of the Advanced Engineering BE expo at Antwerp Expo. A lot of AI talks and nice demos but I talked about another very important topic before you can develop AI solutions. All the AI business cases rely on data, large sets of data. When you, as a manufacturing company, want to start with a digital transformation you will need to take into consideration that you require a strong and scalable data platform. We have defined 4 key principles where your data platform needs to comply, making it become future proof to implement your AI use cases. Thanks at The Beacon | Innovation Community for the opportunity to have a speaking slot at this event. #codit #azure #microsoft #thebeacon #data #dataplatform #advancedengineering #ai #analytics #manufacturing #iiot #industrialiot

39

API vs events, Belgium edition 🇧🇪 Very excited to be speaking at DevDay.be on the 7th of November, I'll be speaking about API vs event driven architectures. The topic is getting some heat with: - the generative AI trend causing business users to seek real time data. Events tend to enable that - new exciting specifications in the API space such as Arazzo to document complex workflows (Erik Wilde 😃) - tech companies struggling to tick the "Loosely coupled teams" capability from DORA / Accelerate If you are a CTO, CIO or engineering leader with an opinion on either or both approaches, I'd be delighted to hear from you to get some extra insights for the talk. #API #Event

55

How can we ensure that we use AI in a truly frugal way? Dive into the three critical questions that we ask to measure AI's efficiency and impact. Read the blog by Claude Le Pape-Gardeux, Data & AI Domain Leader and Jacques Kluska, Expert Data Scientist: http://spr.ly/60439ivrd #FrugalAI #AI #DigitalTransformation

142

1 Comment

If you want to understand why LLMs are not the solution for AGI (Artificial General Intelligence), I highly recommend listening to Lex Fridman podcast with Yann LeCun : Meta AI, Open Source, Limits of LLMs, AGI & the Future of AI T͟L͟D͟R͟: 1. LLMs do not understand the physical world 2. LLMs have not the ability to remember 3. LLMs have no persistent memory 4. LLMs are not able to reason and to plan

27

5 Comments

Finitism in #AI datasets can be mitigated if a more significant portion of the worldwide internet populace can create data, and if there were a method to incentivize these interested contributors to create such data. A big drum roll for Pertin-ant 🐜here... #AI #AGI #LLMs #trainingdata

6

💥 New day, new model 💥 We ran our performance benchmarks on Anthropic Claude 3.5 Sonnet v/s OpenAI's GPT-4o & the results weren't what you'd expect ⏬ Anthropic says Claude 3.5 Sonnet outperforms all competitor models on key evaluations, at twice the speed of Claude 3 Opus and one-fifth the cost. In our benchmarks 3.5 Sonnet & 4o came out neck & neck. Ultimately, you need to run evaluations on your task at hand and measure the quality yourself. 🔍 Data Extraction: Both models identified 60-80% of data correctly, but neither excelled in this task. 📊 Classification: Claude 3.5 Sonnet (72%) outperformed GPT-4o (65%) in mean accuracy. However, GPT-4o led in precision (86.21%), crucial for accurately classifying customer tickets. 🧠 Verbal Reasoning: GPT-4o outperformed Claude 3.5 Sonnet with 69% accuracy, versus 44%. Both models did well on analogy and relationship questions. 📜 Context window: Claude 3.5 Sonnet boasts a larger context window (200K) compared to GPT-4o's 128K 🌪 Latency: Claude 3.5 Sonnet is 2x faster than Claude 3 Opus, but it’s still lags behind GPT-4o when it comes to latency (50% faster Time to First Token) Price and throughput for these models was very similar Want to learn more about how these models stack up? Check out our full analysis on the Vellum blog: https://lnkd.in/dwEcQJe4 Oh and did I say, Claude 3.5 Sonnet is already in Vellum? Sign up for a demo to run your evals & start building!

68

1 Comment

📢 New Case Study Release! I’m excited to present the latest digital leaflet from LACO, showcasing the successful data governance implementation I contributed to at FEDERALE Assurance-Verzekering! 👏 Big shout out and gratitude to Amandine Rouvroy and Pascal Polet 👏 Check out the full case study attached. (also link in comments) #DataGovernance #DataQuality #DigitalTransformation #CaseStudy

15

1 Comment

⭐️ New video release 📺: Your Model _Probably_ Memorized the Training Data Watch as privacy activist and data scientist Katharine Jarmul reveals the hidden truth about deep learning models memorizing training data and explores the legal risks and privacy implications associated with it. 📺 Watch the video on YouTube: https://lnkd.in/dbqvejYX Katharine Jarmul, a leading expert in #DataPrivacy, delivered a talk shedding light on the likelihood of deep learning models memorizing training data. She emphasized that even large models like language and multi-modal models are not immune to this phenomenon. Katharine discussed the active research on deep learning memorization and its implications, highlighting scenarios where memorization could be beneficial and the associated legal and privacy risks. The talk covered various aspects including the mathematical rationale behind model memorization, successful attacks on deep learning models to extract memorized data, and the social and legal consequences of using such data. Katharine also explored potential solutions like differential privacy, federated models, and distillation techniques to mitigate the risks posed by model memorization. Ultimately, the talk aimed to raise awareness about the issues surrounding model memorization and advocate for thoughtful consideration of privacy and security within data science workflows.

8