This week I've been lucky to visit NYC 🏙 to "talk data" with one of our flagship customers. Some thoughts that i'm taking home with me. 1️⃣ How to avoid "shipping the org chart" for platform engineering teams, specially for multi faceted data journeys (e.g. Data Science / MLOps) 2️⃣ What are the limits of the Data Mesh paradigm? Did we go too far with the decentralized approach? 3️⃣ When it comes to applying Generative AI to the "Data world", is it better to construct user surfacing features (e.g. chatbots, code assistants) or use it "under the hood" (e.g. from notebook to pipeline)

Want to be a machine learning engineer? Yep, data science and AI bootcamp. Dreaming of being an AI engineer? You guessed it—data science and AI bootcamp. Aiming to be a data scientist? Say it with me… data science and AI bootcamp. Stay tuned, cohort 9 is coming soon!

🚀 Day 6 of My 100 Days of Computer Vision Challenge 🚀 Today was a deep dive into enhancing the performance of CNN models by focusing on two critical aspects—data shuffling and improving model architecture. Here's what I worked on: 🔄 Discovering the power of shuffling data: Shuffling the training data during each epoch can have a big impact on model performance. It prevents the model from learning patterns based on the order of the data, making it more robust. This simple but powerful technique ensures the model doesn’t overfit or learn irrelevant sequence patterns. 🔧 Exploring options to improve the model: I spent time tweaking different elements of the model—experimenting with layer configurations, activation functions, and optimizers. Every small adjustment helps find that sweet spot where the model performs better without increasing complexity unnecessarily. Day 6 was all about refining and optimizing my CNN to achieve better results. It's incredible to see how small changes can lead to significant improvements. Excited to see where these insights take my model next! 💪 #100DaysOfCode #ComputerVision #DeepLearning #AI #CNN #ModelOptimization #MachineLearning #TechJourney #LearningByDoing #DataShuffling #ModelImprovement

Presentation on "Lagrange Interpolation Polynomial Approximation". It’s a powerful tool in mathematical modeling, with applications that stretch across machine learning, data science, and engineering etc. I had a great time diving deep into how interpolation techniques like Lagrange can bring precision to predictive models and enhance data-driven insights. Looking forward to connecting with others passionate about math in ML, AI and Data Science. #MathInML #LagrangeInterpolation #MachineLearning #DataScience #EngineeringMath

Why reinvent the wheel when you can borrow the blueprint? 🤷🏽 Discover how a seasoned ML/AI Engineer and Data Scientist, Navar Nascimento, simplifies code sharing by building a Computer Vision App—with practical insights on standardizing filters to make life easier for engineers. 📺 Watch now: https://hubs.ly/Q02RDM1Z0 #TechHumor #ComputerVision #DataCulture #DataEngineering #DataScience

🌟 Exciting News in the World of Data Science! 🌟 I'm thrilled to announce the launch of my latest project: Data Science Tutor Chatbot powered by Gemini 1.5 Pro(LLM) and Streamlit. 🤖💬 This project wouldn't have been possible without the assistance of Kanav Bansal sir at Innomatics Research Labs. Key features: 👉 Solves Data science related queries only. Explore the GitHub repository for a closer look: https://lnkd.in/dkPm6qVe #DataScience #AI #MachineLearning #Gemini #Streamlit

LLMs and Generative AI! In the 6th and final session of the AI Education Series for this semester, we will continue with the overview of LLMs and Generative AI and look into a few of the most popular frameworks for training and using LLM models. Register today for this free, virtual education series. https://lnkd.in/gkeUqp4m

Just completed 40 hrs Data Science and AI training at my college!

Just finished the course “Computer Vision for Data Scientists” by Harpreet Sahota! Check it out: https://lnkd.in/dg7Ksq7G #computervision #deeplearning.