THE ML ENGINEER · YEAR ARCHIVE

Newsletter issues from 2025

Newsletter home

#367Dec 2025Karpathy 2025 Year in Review, Text-to-SQL Optimization Learnings, MIT Distributed Systems, Python Data Science Handbook + more 🚀ai-agents · llms · data-engineering #366Dec 2025MLOps 2015-2035; The Past and Future, Andrew Ng AI Career Advice, 1000+ Case Studies LLMOps, Top Python Libraries in 2025 + more 🚀mlops · llms · ml-education #365Dec 2025PyTorch Hardware Acceleration in 2025, Anthropic MCP now in Linux Foundation, OpenAI Enterprise AI Report, AI Eats the World 2025 + more 🚀gpu-compute · ai-agents #364Dec 2025Raschka on DeepSeek v3.2, DeepMind Nobel Prize Documentary, The Race of DuckDB / Polars / Spark, Advent of SadServers 2025 + more 🚀llms · mlops #363Dec 2025Black Forest Labs Launches FLUX-2, Major Supply Chain Attacks, Raschka on Building an LLM, 2025 Advent of Code + more 🚀generative-ai · ml-security · llms #362Nov 2025Google Weather Forecast 2.0, META on GenAI World Models, MCP Protocol UI Extension, DuckDB's Take on the Lakehouse + more 🚀forecasting · generative-ai #361Nov 2025Stanford CS236: Deep Generative Models, Introducing the OWASP Top 10 2025, GenAI World Foundation Models, Synthetic Time Series Foundation Models + more 🚀generative-ai · ml-security · forecasting #360Nov 2025Tabular Foundation Era Begins, AI-Assisted Engineering Report, PyTorch Agent Framework (+RL), Bench for Semantic Query Engines + more 🚀llms · ai-agents #359Nov 2025DORA AI-Dev Report 2025, Chip Huyen on Lenny's Podcast, Europe's EuroLLM Launches, Learning PyTorch the Hard Way + more 🚀llms #358Oct 2025Amazon Releases Chronos-2, Robotics Foundation Models, DeepSeek OCR Released, Ray Joining Linux Foundation + more 🚀forecasting · llms · mlops #357Oct 2025Claude Skills Driving Agentic Innovation, Truths in Software that are False in AI, Karpathy on the Decade of Agents, Blameless Culture in Engineering + more 🚀ai-agents · reinforcement-learning #356Oct 2025A History of Large Language Models, Few Samples Break LLMs, Uber Michelangelo in 2025, Companies Measure AI Productivity + more 🚀llms · ml-security · mlops #355Oct 2025Andrew Ng New Stanford ML Course, Introduction to Multi-Armed Bandits, Designing Agentic Loops, Building Generative Recommendations + more 🚀ml-education · ai-agents · recommender-systems #354Sept 2025650+ GenAI Case Study Learnings, The Top Languages in 2025, MIT on the GenAI Business Divide, Leading in a Room with Experts + more 🚀llms #353Sept 2025Anthropic Outage Post-Mortem, Harvard on LLM National Bias, The State of Devs in 2025, Build Containers from Scratch + more 🚀mlops · llms · ml-education #352Sept 2025Hierarchical Reasoning Models, QWEN3 Towards Efficiency, Multi-Agents for CVE Exploits, Simon Willison Vibe Code Apps + more 🚀llms · mlops · ml-security #351Sept 2025The GPU Compute Compendium, OpenAI on LLM Hallucinations, Claude Code Tutorial From Scratch, Google on Limitation of Embedding Retrieval + more 🚀gpu-compute · llms #350Aug 2025Stanford Study on AI Automation Impact, Python: The Documentary, A Software Eng Agent in 100 Lines, Developer Productivity Metrics Reading List, SpaCy Against LLM Maximalism + more 🚀ai-agents · nlp #349Aug 2025650 Prod MLOps Case Studies, McKinsey on GenAI Minimal Value, Google DeepMind Measuring Environmental Impact, Metadata Mgmt for AI Flows, Beating Djikstra (for Sparse Graphs) + more 🚀mlops #348Aug 2025LeadDev AI Impact Report, MCP Vulnerabilities For Every Dev, Architecting Large Software Sys, Machine Learning Visualised, Best Practice for Agentic Systems + more 🚀ml-security · ai-agents · ml-education #347Aug 2025Atlassian 2025 State of DevEx, GPT-5 Attributes, Pricing + Card, Qwen Image Technical Report, Google DeepMind World Model, Raschka on Qwen 3 From Scratch + more 🚀llms · ai-agents · generative-ai #346Aug 2025StackOverflow Dev Survey, Free Databricks MLOps Course, Google DeepMind World Mapping, Multi-Generation Projects, ACM Transitions to Open Access + more 🚀mlops · llms · ml-research #345Jul 2025Reverse Eng Github's App AI, Raschka's LLM Architectures, State of Homomorphic Encryption, Ray Data, Train and Tune, Writing Toy Software to Improve + more 🚀llms · ai-agents · mlops #344Jul 2025The State of GenAI in 2025, Neural Net Operating System, Chain of Thought Monitoring, Crawling a Billion Pages, Principal Engineering at Amazon + more 🚀ai-agents · ml-education #343Jul 2025Andrew Ng on Building Faster, GetDX: Measuring AI Productivity, Coding Languages by Efficiency, Twitter Releasing Grok Model, Decreased Productivity with AI + more 🚀ai-agents · llms #342Jul 2025Microsoft AI Path to Medical Superintelligence, OpenAI Practical Guide to Agents, McKinsey's Trillion War for Data Centers, Bloom Filters by Example, The Jax ML Scaling Book + more 🚀ai-agents #341Jun 2025Internals of vLLM Inference, Databricks Unified ML Governance, A Deep Dive Into TPUs, Google MultiModal Released, LLM Transformers as Malware + more 🚀mlops · ai-policy · ml-security #340Jun 2025Karpathy on Software Evolution, Google on Securing AI Agents, Anthropic on Building AgenticSys, State of Eng Leadership 2025, Quantum Computing Lecture Notes + more 🚀ai-agents · llms #339Jun 2025Time-Blind Large Visual Models, Unified Metadata Model at Netflix, Mistral Reasoning Model Magistral, Unreasonable Effectiveness of For-Loop in GenAI, Human Quality Text-to-Speech + more 🚀ai-agents · generative-ai #338Jun 2025The State of AI Talent in 2025, Migrations with LLMs at Google, Claudfare's GenAI Development, Apple on Limitations of LLMs, Mistral Release Codestral Embed + more 🚀llms #337Jun 2025Github CEO on AI Code Race, Black Forest Labs Flux.1 Kontext, The Hidden Bloat on ML Systems, OpenAI ChatGPT Academy, Introducing Perplexity Labs + more 🚀ai-agents · generative-ai #336May 2025Datadog OSS Observability ML, Anthropic Releases Claude 4, Upgraded Climate Foundation AI, Mistral Launches Devstral OSS, Microsoft AI Red Teaming Labs + more 🚀ai-agents · llms #335May 2025OWASP Agent Name Service, OpenAI's New Software Engineer, On Complex Systems at Scale, Google on Text-to-SQL models, DeepMind on Coding Agents + more 🚀ai-agents · generative-ai · mlops #334May 2025LegoGPT: New Model on LEGO, How GenAI Sees Accents, Comparison of SotA Image GenAI, Google Measuring Tech Debt, Deep Dive into PyTorch Internals + more 🚀generative-ai · ml-research · mlops #333May 2025OpenAI's Rouge AI Model, The ML Leaderboard Illusion, When ChatGPT broke NLP, Alibaba's Qwen3 Released, Preparing for Engineering Leadership + more 🚀mlops · llms · nlp #332Apr 2025The GenAI Ecosystem in 2025, AI Code Generation for Research, Forecasting Principles & Practice, META AI Multi-Modal Architecture, Google Gemma for Consumer GPUs + more 🚀mlops · ai-agents · forecasting #331Apr 2025Open Source $8 Trillion Economic Value, Google Gemini, Wisk & Veo2, DeepSeek's Distributed File System, TikTok's LLM Reinforcement Learning, Lessons from the Best Programmers + more 🚀generative-ai · llms · reinforcement-learning #330Apr 2025Stanford AI Index Report, Google's Agent2Agent Protocol, Yann LeCun on Future of LLMs, The S in MCP is for Security, DeepMind and GenAI Competition + more 🚀ml-research · ai-agents · llms #329Apr 2025Survey on Top ML Challenges, Understanding ML Theory to Algos, The 13 Software Laws to Live By, META Released LLaMa 4, Reinforcement Learning From Scratch + more 🚀llms · reinforcement-learning · ml-education #328Mar 2025Anthropic on LLM "Biology", Architectural Patterns in Python, Tencent's New Ultra-Large Model, Improving RecSys in LLM Age, OpenAI's Image Gen Models + more 🚀llms · recommender-systems · generative-ai #327Mar 2025Tencent's Image-to-3D AI, Netflix RecSys Foundation Model, Uber's Text-to-SQL AI Systems, LLMs as Better Meme Generators, Top 100 Most Watched Python Talks + more 🚀generative-ai · recommender-systems · llms #326Mar 2025Google releases Gemma 3, Block Diffusion Challenging LLMs, Mistakes in Large Codebases, Probabilistic Artificial Intelligence, The Startup CTO's Handbook + more 🚀generative-ai #325Mar 2025Karpathy's Tips on LLMs, China's New Autonomous AI Agent, Reading Papers for Engineers, MIT Distributed Systems Course, Difussion Training in Micro-Budget + more 🚀ai-agents · llms · data-engineering #324Mar 2025WanX Text-to-Video Model, The State of ML Competitions, Chip Huyen Beyond Agent Hype, CMU Database Systems Course, TikTok's E2E LLM Stack OSS + more 🚀generative-ai · llms #323Feb 2025OWASP Agentic AI Security, Sakana AI Making CUDA Slower, Karpathy Deep Dive on LLMs, AI Code Increase Tech Debt, Bluesky Lossy Timeline / Ranking + more 🚀ai-agents · llms #322Feb 2025Microsoft on GenAI Making Us Dumb, Text to SQL: The Ultimate Guide, Perplexity Deep Research, Anthropic Economic Index, Build Your Own <Anything> + more 🚀generative-ai · ai-agents #321Feb 2025The RLHF Book, Music AI Generation & Lawsuits, Programming Changing as We Know It, New FC Foundation Model from China, Opinions from a Decade in Software + more 🚀llms #320Feb 2025International AI Safety Report, The DeepSeek Illustrated Guide, AI Engineering Pitfalls w Chip Huyen, CMU ML In Production Course, No-Hype DeepSeek Reading List + more 🚀mlops #319Jan 2025DeepSeek AI Taking by Storm, Add RecSys Engine to Your Apps, 7 Most Influential CompSci Papers, LLMOps in Prod Case Studies, OpenAI's New Browser "Operator" + more 🚀llms · data-engineering · ai-agents #318Jan 2025AI Engineer 2025 Reading List, Causal Inference meets Deep Learning, Jensen Huang NVIDIA Keynote, OpenAI’s Economic Blueprint, Papers Every Dev Must Read + more 🚀ml-research · ai-policy #317Jan 2025Chip Huyen on Agentic AI, 10y Review of Anomaly Detection, Modular SVGs with GenAI Models, Efficient Coding with LLMs, Extracting AI models from Apps + more 🚀ai-agents · llms · ml-security #316Jan 2025Databases Year in Review 2024, Crunching Data Anywhere with DuckDB, Learnings from LLMs in 2024, Evolution of SRE at Google, The Quiet Chinese AI Giant + more 🚀data-engineering · llms · mlops