Report · MoneyHarald Kirschner reports that GitHub issues are a massive signal for holding quality when you're on GitHub. They get sometimes even better issues now because of AI and more people with different language backgrounds being able to file very good issues. Previously, engineers actually triaged issues manually. Now AI does the filtering to get out spam, enriching the data, and then translating, in the end assigning area owners for all the issues to raise the signal-to-noise quality. Add a human layer to AI issue triage — once you are in the issue, because the agent will make mistakes and we need that feedback loop.
AI Engineer · How VS Code Went from Monthly to Weekly Releases with AI — Harald Kirschner
Report · MoneyHarald Kirschner reports that the VS Code team started tracking code survival metrics early on — the percent of agent-written code that actually gets committed. GPT 4.1 started at 55% code survival, and Claude Opus 4.6 is now at 86%. The increase shows developer trust and confidence in shipping AI code faster and more AI code generated.
AI Engineer · How VS Code Went from Monthly to Weekly Releases with AI — Harald Kirschner
Report · MoneyHarald Kirschner reports that code review is the next big investment. When they enabled the automatic GitHub Copilot code review, they weren't convinced yet, but over the next months and weeks it massively improved. Now it's a mandatory review every time somebody opens a PR. After the AI review is done, humans will not even look at a PR until all the comments are resolved and addressed. You can dial in the effort a code review does depending on how much risk you have in the repo — low, medium, high effort for code reviews — to dial in the cost benefit.
AI Engineer · How VS Code Went from Monthly to Weekly Releases with AI — Harald Kirschner
Report · MoneyThe initial architecture routes messages from an iPhone or MacBook through WhatsApp to NanoClaw on the Raspberry Pi, requiring a credential registration on NanoClaw so it can pull messages from WhatsApp in the cloud. No LLM inference happens on the Pi — Jeremy uses the cloud for all inference, describing it as 'a big brain over a wire to the cloud' — and he also has a Neo4j database in the cloud alongside the inference.
AI Engineer · I Built a Personal AI Agent on a Raspberry Pi — Jeremy Adams, Neo4j
Creator perspective · MoneyThe final system: a request comes in, a serverless authentication function authenticates the browser, the authenticated browser is provided to the agent runtime, the agent navigates the portal using the skill, calls a deterministic download function, and provides the downloaded file to a verify tool so it can know authoritatively whether its trajectory succeeded or failed. This architecture takes steps out of the model's responsibility and, Meegan argues, can overcome the browser agent conundrum.
: none
AI Engineer · Why 99% Accurate Browser Agents Still Fail — Derek Meegan, Browserbase
Report · MoneyCurrent protocols like MCP are stateless, making sticky sessions difficult. A2A is client-server based, requiring complex dual-role setups for bidirectional communication, and lacks built-in discovery mechanisms, forcing developers to handle planning and infrastructure manually.
AI Engineer · Your Agents Are in Solitary Confinement: Why MCP & A2A Aren't Enough — Vlad Luzin, Band
ReportCurrent AI coding agents (e.g., Cloud Code, Codex) primarily accelerate the production of existing software types rather than expanding software capabilities. They automate minor tasks like small pull requests (~10 lines) but do not create fundamentally new software functions.
high
The a16z Show · AI Can Write Code. Why Isn’t Software Better?
ReportThe host describes the second issue of i-D featuring a baby on the cover, representing Gen Beta, and notes the baby was photographed winking. In lieu of a profile, the host asked important artists, politicians, and figures to give advice to someone born in 2025.
Figma · 8 Designs That Inspired i-D Magazine's Editor in Chief | Source Material
ReportThe host describes the first issue of i-D magazine under their editorship, which featured a nationwide casting call to find an unknown cover star. They found Enza, a senior in high school in suburban Cleveland, Ohio, who was an aspiring model and LARPer. The host and photographer Aiden Zamiri spent three days visiting her, hanging out and seeing her world, then shot her in her environment: her room, her living room, and at her drama club.
Figma · 8 Designs That Inspired i-D Magazine's Editor in Chief | Source Material
Report · MoneyThe people who pay the most are strategic buyers or growth-focused private equity. Founders should target these buyer types.
MicroConf · When to Sell Your SaaS Business: How to Get a 5x Better Offer
ReportAtlas is a world model that performs three core capabilities: generation (creating new worlds from text/image prompts), reconstruction (modeling real-world spaces from photos), and simulation (recreating spaces to simulate robot behavior). It is a rebuilt, more unified architecture compared to its predecessor, Marble.
The a16z Show · World Models, Robotics, and the Future of 3D AI
ReportAtlas can recreate a real-world space in 3D and simulate how a robot might behave in that space. This enables a 'real-to-sim-to-real' workflow for adapting robotics foundation models to specific environments, and is particularly useful for the 'last mile' of training and evaluation.
The a16z Show · World Models, Robotics, and the Future of 3D AI
ReportHodak describes the retinal prosthesis (Prima) as a proof of concept for the brain-as-computer approach, demonstrating that form vision can be restored in blind patients, which had never been done before.
No Priors: Artificial Intelligence | Technology | Startups · From Restoring Sight to Reimagining the Brain, with Max Hodak
ReportHodak reports that in the clinical trial, patients were able to fill in Sudoku puzzles, crossword puzzles, and read books, providing existence proof that the approach works.
No Priors: Artificial Intelligence | Technology | Startups · From Restoring Sight to Reimagining the Brain, with Max Hodak
Recommendation · MoneyThe speaker advises doing a coding evaluation, a writing evaluation, Humanity's Last Exam, and trying to have as many paradigms as possible. He adds that game-based evaluation could be one paradigm of model evaluation you can add into your stack.
AI Engineer · If Your AI Scores 100%, Your Eval Is Wrong - Ramakrishnan Lokanathan, eslams.com
Recommendation · BodyDr. Marius Bakken recommends walking up the same stairs every day and noticing how heavy the legs feel as a really good way to monitor muscle tone day to day, plus springiness on actual sessions — how light you feel each step — as a second signal. He names the stair test, how you feel on the session, and how it evolves as the most important monitoring tools; lactate versus heart rate and pace are more complicated options but not necessary.
Commercial context: The creator presents this as their own offering and directs viewers to get it.
Garage Strength · The Genius Who Solved Endurance Training w/ Dr. Marius Bakken
Recommendation · BodyDr. Marius Bakken recommends splitting the two weekly threshold sessions into intervals: one with short intervals (45 seconds to 1 minute) and one with longer intervals (about 6 minutes), keeping total running time around 25 to 35 minutes — or 25 to 40 minutes depending on level and weekly volume. For longer repeats, take 1 to 1.5 minutes recovery; for one-minute repeats, 30 seconds, limiting recovery more strictly on shorter repeats because short recoveries matter more there. If you have fairly good control of intensity you can do 10-minute repeats; otherwise stick to 6-minute or even 4-minute repeats.
Garage Strength · The Genius Who Solved Endurance Training w/ Dr. Marius Bakken
Recommendation · BodyBakken recommends that the host get a lactate monitor to test during treadmill or track interval sessions, if he uses it properly. A lactate monitor should be used either extensively throughout training or just occasionally to get a feel — the failure mode is using it too little to see patterns. For the host, it will give an idea of where he actually is in his training, but he should not necessarily use it on every session.
Garage Strength · The Genius Who Solved Endurance Training w/ Dr. Marius Bakken
Recommendation · BodyModern running shoes with carbon plates and improved foam are one of the biggest changes in running in recent years. Since 2012 Bakken has been a doctor with the same patient list of about 1,200 patients, many of them runners. Around 2018-19, carbon-plate and improved-foam shoes entered the market more and more, and he observed a big shift in running injuries in his clinical practice; more patients could actually keep running when they used them. If you are getting into running or starting to get into trouble, get modern running shoes with a carbon plate or modern foam. If you are having trouble, switch shoe brands to see if it benefits you more, because the upside is quite big.
Consequential creator advice requires independent verification before Today can promote it.
Garage Strength · The Genius Who Solved Endurance Training w/ Dr. Marius Bakken
Recommendation · MoneyThe team endorses five areas: (1) standardize evaluation infrastructure with reusable evaluation frameworks, metrics, and judges calibratable with human feedback; (2) automate feedback loops so systems continuously collect execution traces, detect recurring failures, and perform root cause analysis; (3) scale automated optimization including prompt optimization and agent reflection, where candidate updates can be generated and tested in a repeatable way; (4) validation and trust — any automated improvement needs evidence it works, via robust holdout evaluations and regression testing; (5) democratize RSI across teams with reusable components and self-service info so teams don't build an entire RSI pipeline themselves.
AI Engineer · An Agent That Fixes Its Own Mistakes - Saloni Gandhi & Ziwei Guo, LinkedIn