You're in an ML Engineer interview at Netflix.The interviewer asks:"You’ve trained a new recommendation model.How do you make sure it’s ready to replace the old one?"You reply: "I’ll compare metrics on validation and test sets."Interview over.Here’s what you missed:The issue is that, despite rigorously testing an ML model locally (on validation and test sets), it could be a terrible idea to instantly replace the previous model with the new model.This is because it is difficult to replicate the exact product ...

AI Agent Framework - Google launched ADK, an open-source framework for building, orchestrating, evaluating, and deploying production-grade Agentic systems [1] - Google ADK is compatible with MCP (for connecting to external tools), A2A (for connecting to other agents), and AG-UI (for connecting to users) [1] AG-UI Protocol - AG-UI is an open-source protocol enabling agents to collaborate with users [2] - AG-UI facilitates a bridge between a backend AI agent and a full-stack app [2] - Connecting an agent to a React frontend using CopilotKit involves defining the agent with ADK and connecting it to the frontend [2]

If you found it insightful, reshare it with your network.Find me → @_avichawlaEvery day, I share tutorials and insights on DS, ML, LLMs, and RAGs.Avi Chawla (@_avichawla):Google did it again!First, they launched ADK, a fully open-source framework to build, orchestrate, evaluate, and deploy production-grade Agentic systems.And now, they have made it even powerful!Google ADK is now fully compatible with all three major AI protocols out there: https://t.co/nMkcyLEhyl ...

GitHub repo: https://t.co/FfVx9UU6d3(don't forget to star it ⭐ ) ...

Google did it again!First, they launched ADK, a fully open-source framework to build, orchestrate, evaluate, and deploy production-grade Agentic systems.And now, they have made it even powerful!Google ADK is now fully compatible with all three major AI protocols out there:- MCP: To connect to external tools- A2A: To connect to other agents- AG-UI: To connect to users.AG-UI is the newest addition, which is an open-source protocol that enables agents to collaborate with users.They worked with the AG-UI team t ...

Technology & Performance - LLM (Large Language Model) inference speed is affected by the use of KV caching [1] - The tweet shares a resource comparing LLM inference speed with and without KV caching [1]

If you found it insightful, reshare it with your network.Find me → @_avichawlaEvery day, I share tutorials and insights on DS, ML, LLMs, and RAGs.Avi Chawla (@_avichawla):LLM inference speed with vs. without KV caching: https://t.co/ReVMJMteKa ...

The visual explains the underlying details of KV caching.I also wrote a detailed explainer thread on KV caching a few months back, if you want to learn more.Check below👇 https://t.co/e4KILO0cEeAvi Chawla (@_avichawla):KV caching in LLMs, clearly explained (with visuals): ...

Inference Optimization - LLM 推理速度对比，有无 KV 缓存 [1]

Model Training Strategy - The initial approach of capturing user images and training a binary classifier for face unlock is flawed due to the need for on-device training and the difficulty of obtaining "Class 0" samples [1][2] - A Siamese Network trained via Contrastive learning offers a more suitable solution for face unlock systems [2] - Contrastive learning maps data points to a shared embedding space, where low distance indicates similarity and high distance indicates dissimilarity [3] - The system creates a dataset of face pairs, labeling pairs of the same person as 0 and different people as 1, then trains a supervised model [3] - A neural network generates embeddings for each image, and the distance between embeddings is minimized for similar faces and maximized for dissimilar faces using contrastive loss [4] - The contrastive loss function, L = (1-y)*D^2 + y*max(0, margin-D)^2, guides the model to produce low distances for similar inputs and high distances for dissimilar inputs [5] Face Unlock System Implementation - During setup, the user's facial data generates a reference embedding, and subsequent unlocks compare new embeddings against this reference embedding without further training [6] - New identities can be added by creating additional reference embeddings [6] - During unlock, the incoming user's embedding is compared against all reference embeddings [7]