Taichi Nakashima

exe.dev - AI時代のVM Hostingサービス

書いた / exe.dev - AI時代のVM Hostingサービス www.deeeet.com/writing/exe-...

Agentic AI Foundation (AAIF) Agentic AI Foundation (AAIF) is the neutral and open foundation built on transparency, collaboration, and standardization to advance the public interest in agentic AI innovation.

Agentic AI Foundation (AAIF) aaif.io

Six New Tips for Better Coding With Agents I’m hanging out in Sydney with my esteemed co-author and co-conspirator Gene Kim today; we flew in to conduct Vibe Coding workshops and…

Six New Tips for Better Coding With Agents steve-yegge.medium.com/six-new-tips...

Why I Ignore The Spotlight as a Staff Engineer lalitm.com/software-eng...

Transparent Leadership Beats Servant Leadership entropicthoughts.com/transparent-...

Why is quality so rare? - Linear The following is adapted from a keynote talk I delivered at Config in May 2025. I've edited it slightly for clarity, but wanted to share these thoughts on craft, quality, and the choices we make as…

Why is quality so rare? - Linear linear.app/now/why-is-q...

Introducing AWS Transform custom: Crush tech debt with AI-powered code modernization | Amazon Web Services Modernize applications at scale with this new AI-powered service that learns your organization's patterns, automates transformations across repositories, and cuts execution time by up to 80% through…

Introducing AWS Transform custom: Crush tech debt with AI-powered code modernization | Amazon Web Services aws.amazon.com/jp/blogs/aws...

Building an AI-Native Engineering Team developers.openai.com/codex/guides...

The 2025 Go Cryptography State of the Union I delivered my traditional Go Cryptography State of the Union talk at GopherCon US 2025 in New York. It goes into everything that happened at the intersection of Go and cryptography over the last…

The 2025 Go Cryptography State of the Union words.filippo.io/2025-state/

How we built a 130,000-node GKE cluster | Google Cloud Blog Learn about the architectural innovations we used to build a 130,000-node Kubernetes cluster, and the trends driving demand for these environments.

How we built a 130,000-node GKE cluster | Google Cloud Blog cloud.google.com/blog/product...

Software 2.0 I sometimes see people refer to neural networks as just “another tool in your machine learning toolbox”. They have some pros and cons, they…

Software 2.0 karpathy.medium.com/software-2-0...

Jason Spielman • Designing NotebookLM Designer, builder, and visual storyteller. Now building Huxe. Previously led design on NotebookLM and contributed to Google AI projects like Gemini and Search. Also shoot photo/video for brands like…

Jason Spielman • Designing NotebookLM jasonspielman.com/notebooklm

Experimenting with local LLMs on macOS A developer's guide to downloading and running LLMs on macOS, for experimentation and privacy.

Experimenting with local LLMs on macOS blog.6nok.org/experimentin...

research.trychroma.com/context-rot

What are Forward Deployed Engineers, and why are they so in demand? Startups and scaleups are on a hiring spree for a software engineering role pioneered by Palantir. A deepdive into this role, and why FDEs are so popular in 2025

What are Forward Deployed Engineers, and why are they so in demand? newsletter.pragmaticengineer.com/p/forward-de...

iPhone DevOps - ultimate edition I wrote earlier about my dream of developing “single-handedly” on an iPhone. Then, I wrote some more about it. The dream still lives strongly! I am now coding single-handedly in any language on my…

iPhone DevOps - ultimate edition clearsky.dev/blog/iphone-...

Why LLMs Can't Really Build Software - Zed Blog From the Zed Blog: Writing code is only one part of effective software engineering.

Why LLMs Can't Really Build Software - Zed Blog zed.dev/blog/why-llm...

I tried Servo, the undercover web browser engine made with Rust Servo was supposed to be Firefox's future. Now it's an independent effort to make a fast and secure web browser engine.

I tried Servo, the undercover web browser engine made with Rust www.spacebar.news/servo-underc...

Vibe code is legacy code Updates and articles from the Val Town team

Vibe code is legacy code blog.val.town/vibe-code

LLM Embeddings Explained: A Visual and Intuitive Guide - a Hugging Face Space by hesamation This app explains how language models transform text into meaningful representations through embeddings. It provides a visual guide to help you understand traditional and modern language model tech...

LLM Embeddings Explained: A Visual and Intuitive Guide huggingface.co/spaces/hesam...

Hierarchical Reasoning Model Reasoning, the process of devising and executing complex goal-oriented action sequences, remains a critical challenge in AI. Current large language models (LLMs) primarily employ Chain-of-Thought…

Hierarchical Reasoning Model arxiv.org/abs/2506.217...

Tolerating full cloud outages with Monzo Stand-in It all starts with our personal current account

Tolerating full cloud outages with Monzo Stand-in monzo.com/blog/tolerat...

How I keep up with AI progress (and why you must too) - nilenso blog Atharva Raykar ...

How I keep up with AI progress (and why you must too) - nilenso blog blog.nilenso.com/blog/2025/06...

Introduction | LLM Inference Handbook A practical handbook for engineers building, optimizing, scaling and operating LLM inference systems in production.

LLM Inference in Production bentoml.com/llm/

I used o3 to profile myself from my saved Pocket links noperator.dev/posts/o3-poc...

The Prompt Engineering Playbook for Programmers Turn AI coding assistants into more reliable development partners

addyo.substack.com/p/the-prompt...

Swift at Apple: migrating the Password Monitoring service from Java Swift is heavily used in production for building cloud services at Apple, with incredible results. Last year, the Password Monitoring service was rewritten in Swift, handling multiple billions of requ...

Swift at Apple: migrating the Password Monitoring service from Java swift.org/blog/swift-a...

Journey to 1000 models: Scaling Instagram’s recommendation system In this post, we explore how Instagram has successfully scaled its algorithm to include over 1000 ML models without sacrificing recommendation quality or reliability.  We delve into the intricacies…

Journey to 1000 models: Scaling Instagram’s recommendation system engineering.fb.com/2025/05/21/p...

Vision Language Models (Better, faster, stronger) We’re on a journey to advance and democratize artificial intelligence through open source and open science.

huggingface.co/blog/vlms-2025

Design Token-Based UI Architecture Describing established patterns for organizing design tokens and practical approaches for automation

martinfowler.com/articles/des...