Recommendation · MindHuberman recommends deliberately slowing perceived time — watching an espresso pour with high temporal detail, or lying in the yard watching clouds — while noting you cannot sustain high frame-rate attention all day.
Andrew Huberman · Access the Sublime & Live Your Best Life | Robert Greene
Recommendation · MoneyDon't just obsess over your own product as a power user — scale that obsession with AI. Robby Stein built an internal tool using the anti-gravity system that looks at all potential questions users might ask. Use the product as an agent: go to the site, type in a question, go to the AI, ask a follow-up, record the experience, screenshot it, and evaluate against a comprehensive rubric. Evaluate whether the model renders math and LaTeX well or emits weird characters on homework-style questions. Check whether visual queries like bioluminescence get a visual treatment in the response rather than just text. The team has invested in agents that can figure out when stuff is broken or off spec and increasingly fix it themselves, creating a feedback loop. The days of needing a QA team or PMs obsesss
hedged
Lenny's Podcast · What it takes to be a top PM today | Robby Stein (Google Search)
Recommendation · MoneyUse PR news wires even though PR specialists advise against them, because it works for Ajelix. You always control the narrative: you write the press release with all the features and explanations you want customers to see. When a customer searches for your brand in ChatGPT, it takes information from press release sources.
MicroConf · SEO for AI Search: How Ajelix Grew to 350K Signups Without Ads
Recommendation · SocialThe host recommends asking yourself what the situationship partner has actually done, noting that you won't be able to name anything concrete.
Call Her Daddy · The Situationship Survival Guide
RecommendationNew rule: when a column gets re-tagged as sensitive, search the code base for anyone still reading it from the copy that's about to lose it.
AI Engineer · 60% of Our Data PRs Now Merge With No Human Review - Chetas Joshi, Robinhood
Recommendation · MoneyThe speaker advises that in the mid-term, production cannot wait — without the right amount of eval structure, organizations fail to deliver the promise in production that they make with the LLM. He cites the MIT Gen AI Divide report, which found 95% of organizations are not able to get any returns because they are not delivering the promise in production that they tried to make with the LLM.
AI Engineer · If Your AI Scores 100%, Your Eval Is Wrong - Ramakrishnan Lokanathan, eslams.com
Recommendation · BodyBakken's standing advice for people getting into running or starting to have trouble is to try modern running shoes with a carbon plate or modern foam. He also recommends switching shoe brands to see if a different brand benefits you more. He observes that since roughly 2018-19, carbon-plate and improved-foam shoes reduced running injuries and let more patients keep running in his clinical practice of about 1,200 patients.
Consequential creator advice requires independent verification before Today can promote it.
Garage Strength · The Genius Who Solved Endurance Training w/ Dr. Marius Bakken
Recommendation · BodyBakken recommends doing more marathon-specific work with longer, more durable sessions from about 6 to 12 weeks before the marathon, limited to those marathon-specific weeks rather than year-round. Before that block, he says you can use the math from his book to build a very good base.
Garage Strength · The Genius Who Solved Endurance Training w/ Dr. Marius Bakken
Recommendation · MoneyWhen adding intelligence to gateway connectivity, think about end-to-end user identity connected to the data; for orchestration, think about context assembly, memory, and context construction; at the memory-core/database layer, extend into ontology, semantic layers, and memory.
AI Engineer · Everyone Has the Same AI Models. Here's What Sets Agents Apart - Richmond Alake, Oracle
Recommendation · MoneyWhen information comes in, there are different types of memory, and you want to route it effectively and cheaply, which can be done with decision models; normally memory routing would be done using some code, but with decision models it becomes very easy and very cheap.
AI Engineer · Everyone Has the Same AI Models. Here's What Sets Agents Apart - Richmond Alake, Oracle
Recommendation · MoneyZach Lloyd recommends building a private benchmark suite by taking any past run from your system and adding it to a benchmark set — starting by hand with a set of small, medium, and hard representative tasks. He also notes it is possible to have the factory itself look at prior sessions and identify interesting ones for the benchmark set, since everything is stored in code. He frames the guiding question as 'What's the best model for me to use for front end tasks?' and the curation framing as 'this is a representative thing that I did in my factory that could be useful to measure'.
AI Engineer · How to Build a Software Factory That Improves Itself - Zach Lloyd, Warp
Recommendation · MoneyZach Lloyd recommends letting engineers put work into the factory through the tools they already use — Slack, task trackers like Linear or Jira, and GitHub PR comments — so all work flows into a centralized location with a centralized view and everything gets recorded over time. He notes that forcing new interfaces creates friction, and that engineers responding to a GitHub PR comment should have the same context as through any other channel.
AI Engineer · How to Build a Software Factory That Improves Itself - Zach Lloyd, Warp
Recommendation · MoneyDaksh Gupta recommends building a living wiki that acts as a world model of how your company works, incorporating codebase knowledge, production telemetry, logs, roadmap, documentation, and tickets, and using it to supply context to agents.
AI Engineer · How to Build an AI Code Review Agent - Daksh Gupta, Greptile
Recommendation · MoneyDaksh Gupta describes how Greptile produces an internal knowledge base where they essentially check across the entire code base all the different parts, catalog them, and then index them — for example, 'this is authentication, here's how it works, here are the flows, here's a diagram of the flows, here's every file that's involved'. He notes that detecting an auth bypass requires a pretty concrete understanding of whether you're in an authenticated environment when you're in that part of the code base or not, which is why codebase context matters.
AI Engineer · How to Build an AI Code Review Agent - Daksh Gupta, Greptile
Recommendation · MoneyFeed candidate screen-recording videos and transcripts into a ranking agent and have it stack-rank the candidates. Ryan Carson finds a very intelligent model is pretty good at understanding how good a candidate is, so watching all the videos yourself is optional if you have time.
hedged
AI Engineer · How One Founder Manages 20 AI Coding Agents at Once - Ryan Carson, Untangle
Recommendation · MoneyAsk agent-operator candidates to record a screen capture of themselves shipping a feature PR on an app they already maintain, with no instruction or guidance. This unpaid task is the entry gate to the hiring funnel — candidates don't get paid for it.
AI Engineer · How One Founder Manages 20 AI Coding Agents at Once - Ryan Carson, Untangle
Recommendation · MoneyPay finalists a fixed fee for 16 hours of their time on a real project. Give each finalist a seat on Devin (AI coding agent) and ask them to get a real PR ready for merge.
AI Engineer · How One Founder Manages 20 AI Coding Agents at Once - Ryan Carson, Untangle
Recommendation · MoneyThe OptimizeAnything API requires the set of problems you care to solve along with an evaluator/fitness function that returns a score plus any available domain-specific side information; calling optimize_anything with these returns an optimized solution.
AI Engineer · Beating RL With Reflection: GEPA and Optimize Anything — Lakshya A. Agrawal, GEPA
Recommendation · MoneyRoman Ugarte describes moving toward a design where the bot just starts working and sends progressive updates while internal mechanics like tool calls and chain-of-thought stay hidden, and focusing on making the core task just work rather than adding dropdown menus and bells and whistles.
hedged
Lenny's Podcast · How we built Grok Bot in a month | Roman Ugarte (SpaceXAI)
Recommendation · MoneyRoman Ugarte argues against adding new tabs for each new form factor into an existing product: it feels cluttered, reflects a 'shipping your org chart' style, and users react negatively to multiple visions sharing a screen. He also notes existing products can carry intimidating brand associations for non-technical users and small paper cuts for non-core use cases.
hedged
Lenny's Podcast · How we built Grok Bot in a month | Roman Ugarte (SpaceXAI)