Inspired by Andrej Karpathy's #microGPT, here are eight implementations of the same tiny GPT language model, each version teaching something different about how neural networks are built, trained, and run: github.com/danilop/micr...

Here are the slides for my session today! Context pressure with MCP: Patterns and trade-offs danilop.s3.eu-west-1.amazonaws.com/Presentation...

Hey, I’ll be speaking at #MCPconference London on 11–12 February. My session: Context pressure with MCP: Patterns and trade-offs There's a free ticket for my network: luma.com/mcplondon202... Reach out if you're there! I look forward to the tech chat :)

📰🚨 AWS Weekly Roundup: Kiro CLI latest features, AWS European Sovereign Cloud, EC2 X8i instances, and more (January 19, 2026) #AWSNewsBlog #AWSCommunityDay #reInvent #AWSBuilderCenter #SSACommunity

Here's my AWS Weekly Roundup 👉 AWS Lambda for .NET 10, AWS Client VPN quickstart, Best of AWS re:Invent, and more (January 12, 2026) ift.tt/iwO2SgI #AWS #News #Serverless #VPN

I had a problem. I had accumulated too many MCP servers over time. So I asked Kiro to improve its own configuration and group tools into Powers that are loaded in context only when needed. Self-optimization is a new pattern. dev.to/aws/how-i-us... #AWS #Kiro #DevEx #AI

Hey, Amazon Science is on Bluesky! Welcome :) @amazon-science.bsky.social

Chain-of-Draft (CoD) prompting is a technique that compresses reasoning steps to 5 words or less while maintaining accuracy. #AWS #AI

More details on Mantle, our next-generation inference engine for Amazon Bedrock. Designed with zero operator access (ZOA) from the ground up—no SSH, no Session Manager, no serial consoles. There's no mechanism for any AWS operator to access customer prompts or completions. #AWS #AI #Security

Happy 2026! If learning AI and cloud computing is on your list of resolutions this year, what about starting with an AWS Free Tier account? #CloudComputing #AI #AWS #LearnToCode #TechEducation #DevCommunity #2026Goals

Just got my first (smart) telescope and got these – 20’ each 🤩

Happy holidays from #Rome 🚶🏻‍♂️📷

dev.to/aws/explorin... The OpenAI API format is what most are used to. If you're wondering what switching to Amazon Bedrock would actually look like, the barrier is lower than you might expect. Project Mantle provides OpenAI-compatible endpoints and the same SDK, just a different base URL. (1️⃣/3️⃣) 🧵

Amazon Bedrock now has OpenAI-compatible APIs through Project Mantle. I built a CLI to explore them and learned some interesting things about the two different approaches. 🧵 dev.to/aws/explorin... #AWS #AI #OpenAI #Python

Heavy downside at CloudWatch: Managing logs at scale usually sucks. With Unified Data Management, this may be solved!

RUM is a heavily overlooked feature of CloudWatch, even though it's great. Now it finally supports mobile apps too! 🕵️‍♂️

Inspired by the general availability of Amazon S3 Vectors, I used it with Nova Multimodal Embeddings to build "SemStash", a semantic storage for humans and AI agents: dev.to/aws/building... #AWS #AI #Storage #Python #MCP

Inspired by the general availability of Amazon S3 Vectors, I used it with Nova Multimodal Embeddings to build "SemStash", a semantic storage for humans and AI agents: dev.to/aws/building... #AWS #AI #Storage #Python #MCP

Amazon Bedrock now supports Responses API on new OpenAI API-compatible service endpoints 👉 Responses API support is available today starting with OpenAI's GPT OSS 20B/120B models, with support for other models coming soon go.aws/447OY2F #AWS #AI #OpenAI #reInvent #AWSreInvent #DevEx #SDK

Strands Agents now speaks TypeScript: A Side-by-Side Guide 👉 This article compares the Python and TypeScript implementations side by side, building a practical example along the way dev.to/aws/strands-... #OpenSource #AI #Python #TypeScript

Announcing TypeScript support in Strands Agents (preview) buff.ly/Y8bWvVR #AI #OpenSource

Amazon Bedrock adds 18 fully managed open weight models 👉 Including the new Mistral Large 3 and Ministral 3 models go.aws/3XuL5kE #AWS #AI #reInvent #AWSreInvent

Build multi-step applications and AI workflows with AWS Lambda durable functions buff.ly/6tM27YH #AWS #Serverless #AI #reInvent #AWSreInvent

Amazon S3 Vectors now generally available with increased scale and performance 👉 Reduce the total cost of storing and querying vectors by up to 90% when compared to specialized vector database solutions go.aws/3XuL2oY #AWS #Storage #AI

Introducing Amazon Nova 2 Sonic 👉 Our new speech-to-speech model for conversational AI New languages, improved speech understanding, asynchronous tool calling , and more. buff.ly/Dg3RMpr #AWS #AI #reInvent #AWSreInvent

Amazon Bedrock AgentCore adds quality evaluations and policy controls for deploying trusted AI agents – Plus episodic functionality in AgentCore Memory and bidirectional streaming in AgentCore Runtime buff.ly/t84SD0Z #AWS #AI #reInvent #AWSreInvent

Introducing Amazon Nova Forge: Build your own frontier models using Nova 👉 Start development from early model checkpoints, blend your datasets with Amazon Nova-curated training data, and host custom models securely on AWS. buff.ly/8b5Z3lJ #AWS #AI #reInvent #AWSreInvent

Build reliable AI agents for UI workflow automation with Amazon Nova Act, now generally available buff.ly/Jb6ApsO #AWS #AI

Introducing Amazon Nova 2 Lite 👉 A fast, cost-effective reasoning model for everyday workloads offering industry-leading price performance to build capable, reliable, and efficient agentic-AI applications buff.ly/eJdpOo3 #AWS #AI #reInvent #AWSreInvent

FYI – We’re rounding up the most exciting and impactful announcements from AWS re:Invent 2025 here: ift.tt/I40LBk8 #AWS #News #AWSreInvent #reInvent

Introducing flat-rate pricing plans with a monthly price 👉 With no overage charges, regardless of whether your website or application goes viral or faces a DDoS attack. #AWS #CDN

Introducing flat-rate pricing plans with a monthly price 👉 With no overage charges, regardless of whether your website or application goes viral or faces a DDoS attack. #AWS #CDN

This will help you answer questions such as “I want to call a specific AWS service operation, which IAM permissions do I need?” buff.ly/T8H60Aj #AWS

On my way to the second AWS AI In Practice London meetup! buff.ly/OOxF094 #AWS #AI #GenAI #London #AWScommunity

Join me and hundreds of other developers at AWS Dev Day Stockholm on November 19! I'll share my learning on how automated reasoning can help reduce the risk of AI hallucinations. Register today: buff.ly/Ge4L6hA #awsdevday

Transform your MCP architecture: Unite MCP servers through AgentCore Gateway 👉 A single point of control for routing, authentication, and tool management. including semantic discovery of tools across different target types ift.tt/JzSC7wp #AWS #AI #GenAI #MCP

Tired of manually converting Python projects from "requirements.txt" to modern "pyproject.toml" to use with "uv"? I built a tool that does it in one command. dev.to/danilop/mode... #Python #DevTools #DevEx #OpenSource #UV

Tired of manually converting Python projects from "requirements.txt" to modern "pyproject.toml" to use with "uv"? I built a tool that does it in one command. dev.to/danilop/mode... #Python #DevTools #DevEx #OpenSource #UV

Building AI agents has become easier 👉 Iterate faster with Amazon Bedrock AgentCore Runtime direct code deployment buff.ly/FRHe6YU #AWS #AI #GenAI #Python #Serverless

Love this! Jupyter Deploy 👉 Create a JupyterLab application with real-time collaboration in the cloud in minutes (Jonathan did all the work here, I just helped the post slip through the door) buff.ly/rihvfAd #OpenSource #DevEx

Quick update: Nova Multimodal Embeddings now supports batch inference, allowing to convert large volumes of content into embeddings more efficiently. Instead of sending individual requests for each, you can send multiple items in a single request, reducing API overhead.

Reduce CAPTCHAs for AI agents browsing the web with Web Bot Auth (Preview) in Amazon Bedrock AgentCore Browser 👉 Web Bot Auth is a draft IETF protocol that gives agents verifiable cryptographic identities. buff.ly/zT38oBP #AWS #AI #GenAI #Security

Never Forget a Thing: Building AI Agents with Hybrid Memory Using Strands Agents A proof-of-concept memory system that maintains conversation flow through summarization while preserving the ability to retrieve exact historical messages when needed. dev.to/aws/never-fo...

ICYMI – Introducing Chronos-2: From univariate to universal forecasting In-context learning enables Chronos-2 to solve forecasting tasks with an arbitrary number of dimensions in a zero-shot manner.

Never Forget a Thing: Building AI Agents with Hybrid Memory Using Strands Agents A proof-of-concept memory system that maintains conversation flow through summarization while preserving the ability to retrieve exact historical messages when needed. dev.to/aws/never-fo...

Amazon Nova Multimodal Embeddings: State-of-the-art embedding model for agentic RAG and semantic search 👉 The first unified embedding model that supports text, documents, images, video, and audio to enable crossmodal retrieval with leading accuracy. Read more in my post:

Amazon Nova Multimodal Embeddings: State-of-the-art embedding model for agentic RAG and semantic search 👉 The first unified embedding model that supports text, documents, images, video, and audio to enable crossmodal retrieval with leading accuracy. Read more in my post:

What if you could simulate human behavior before making big decisions? 🤖 Fascinating Stanford webinar on AI agent simulations — how “digital people” built on LLMs could transform research, management, and policy.

Walking in #London #Autumn #Fall 🚶‍♂️📷

Our latest videos are out! 1️⃣ From Prototype to Production with Strands Agents & Amazon Bedrock AgentCore - @danilop.bsky.social 2️⃣ My AI Coding Partner: Building the London Improv Calendar on AWS - Diana Ionita youtube.com/c/AWSUserGro... #AWS #AI #AmazonBedrock #Agents #Serverless #AWSCommunity