Creator perspective · BodyDr. Morin explains that when running fast, good athletes have ground contact times around 100-150 milliseconds, so sprinters need not only a high level of strength in the ankle and foot, but also the ability to express that strength within a super short time window. Some athletes have high maximum strength (e.g., strong plantar flexion iso test) but still collapse at high speed because they only have strength when they have time to express it, while others who are not strong in the maximum strength sense can be super stiff and collapse less because they can produce force within 100-120 milliseconds.
Garage Strength · Why Most Athletes Train Speed Wrong w/ Sprint Scientist Dr. Jean-Benoit Morin
ReportMost of what E2B does is start from everything already running: with templates, you start your processes and they are already running when you come back to the sandbox. The approach is to start processes first and then save the disk state in RAM.
AI Engineer · How I Learned to Stop Worrying and Love the Sandbox — Matt Brockman, E2B
ReportAn E2B template is a snapshotted VM: start a VM, run processes, then save it as a template so others can resume where you left off. The workshop made roughly 200 copies of a template so every attendee starts where the presenter resumed.
AI Engineer · How I Learned to Stop Worrying and Love the Sandbox — Matt Brockman, E2B
Principle · MoneyValuable skills are the rituals nobody wants rediscovered — knowledge that is not glamorous and deeply embedded in your company. Examples: the exact sequence of scripts and feature flags for a nice rollback, or the exact wording to use with a pesky enterprise customer about to churn. If agents try to rediscover this from scratch they will likely fail, and that failure can be really costly.
AI Engineer · From 36% to 100%: How Self-Improving Agents Write Their Own Skills — Rafal Wilinski, Runlayer
Creator perspective · MoneyInstalling community skills via npx without vetting installs prompt injections, root-requiring scripts, and someone else's API key alongside the skill; RunLayer scans thousands of skills daily and finds 'really weird stuff' inside.
AI Engineer · From 36% to 100%: How Self-Improving Agents Write Their Own Skills — Rafal Wilinski, Runlayer
Principle · MindThe ten life elements most people call important are community, romance, parenting, family, recreation, friends, career, physical health, spiritual health, and personal growth; life is zero-sum with time, so an hour invested in one takes an hour from another, and assuming 12 focused hours a day you cannot give each element even an hour and 10 minutes.
The hour math is a hypothetical thought experiment.
Alex Hormozi · It took me 37 years to realize what I'll tell you in 18 minutes.
Creator perspective · MoneyArik describes how trust in AI agents comes with time as users see the agent's precision. In Guitar's user base, users first see precise CI failure analyses uncovering actual bugs and set up rules to block on issues; then they try automated fixes and PR creation; seeing that precision, they unlock auto fix; they add rules to auto-approve PRs touching non-critical code or certain change types; and the next step is having the system unmerge and fix merge conflicts. Full autonomy doesn't come immediately; it comes with using the system and building trust that the agent is precise. When comfortable, users have set up enough rules and guardrails for full autonomy.
described
AI Engineer · Redesigning How Software Gets Built With AI Agents — Sonar & McKinsey Panel
Creator perspective · MoneyGoodwin explains that for iso-intelligence, making mixture-of-experts models sparser (from 1-in-16 to 1-in-128 or 1-in-256) saves a ton of flops. But when serving such sparser MoE models, today's HBM-based GPUs/XPUs become bandwidth-bottlenecked and serve the models at low MFU (flops utilization). As you scale the memory-bandwidth frontier, you get to conserve flops — reducing the flops used to reach a given level of intelligence, which multiplies global throughput.
numeric sparsity figures flagged as possible transcription errors
No Priors: Artificial Intelligence | Technology | Startups · Frontier Chips for Frontier AI Labs, with Walter Goodwin, Founder/CEO of Fractile
Report · MindLiemandt argues that 80-85% accuracy on basic skills is not mastery and that students should nail the basics before moving to advanced material, using a sports analogy: a coach works on ball handling when a player loses the ball 20% of the time, rather than dunks and alley-oops.
Andrew Huberman · How to Accelerate Learning & Improve Education | Joe Liemandt
Creator perspective · MindLiemandt describes Bloom's famous 1980s Two Sigma finding: with an individual tutor to mastery level, kids can learn at two sigma (top 2%). He emphasizes that one sigma comes from the tutor and one from mastery, that mastery is the part everybody forgets (a non-mastery-based tutor won't get students to top 2%), that the claim assumes high levels of motivation, and that you need both individualized and mastery-level instruction to get that performance.
Andrew Huberman · How to Accelerate Learning & Improve Education | Joe Liemandt
Principle · MindJoe Liemandt argues that if you give a student a lesson at their level, they can learn it — a seventh grader who needs fourth-grade material cannot learn seventh-grade lessons but can learn the fourth-grade lesson. US Common Core is designed so that everyone can learn it — the conceptual leaps are not that big. With material delivered at the right level, you can get everybody through eighth grade, and delivering lessons at the student's level solves the IQ issue of whether they can learn the material. After level-appropriate lessons, the student still needs to actually do the work — go to math and grind through the lessons.
Andrew Huberman · How to Accelerate Learning & Improve Education | Joe Liemandt
Creator perspective · MindLiemandt argues that moving a student to the next grade level after finishing fourth grade at 80% mastery is wrong because the problems compound, and describes that in America students scoring at a B level (80%) are moved on to advanced material.
Andrew Huberman · How to Accelerate Learning & Improve Education | Joe Liemandt
Creator perspective · MindLiemandt diagnoses the current education system's core problem as delivering eighth graders eighth grade content, because an eighth grade teacher by law must stand up and deliver eighth grade content. In any classroom there are kids who need seventh, sixth, or fifth grade material, and because of COVID some who need fourth and third grade material due to learning loss. Some kids never learned to read in third and fourth grade and are now being told eighth grade content, which is why the curves are going down. He also blames the removal of basics like memorizing multiplication tables and phonics-based science of reading as lowered standards that kill kids in higher grades; he believes (uncertainly) that educators didn't realize how important these basics were.
partial (rationale for removing basics is uncertain)
Andrew Huberman · How to Accelerate Learning & Improve Education | Joe Liemandt
Creator perspective · MoneyMichael Lee argues the only natural way to partner with world-class organizations, bring frontier engineering to bear, and build a shared platform across portfolio companies is through a holding company. A holding company aligns the duration of capital with the sustained investment and operational commitments required to drive transformation end-to-end, creates an organization culturally oriented around forging market leaders rather than deploying capital, and enables using retained earnings to support existing companies and net new investments.
creator's argument for his own model
No Priors: Artificial Intelligence | Technology | Startups · Re-Founding Incumbents for the AI Era with Sequence Holdings Co-Founder and CEO Michael Lee
Creator perspective · MoneyVarun argues models are there to please you — push hard enough and they will agree with you in every direction — and are designed to give coherent answers but not causality. For production incidents, the only thing you're looking for is a causal chain of evidence, like a detective: the steps that led to the incident, so you fix the right issue instead of creating a patch. Solving production reasoning requires causal reasoning built into the system, not just coherent answers.
AI Engineer · The 6 Pillars of an Agentic Harness for Production — Varun Krovvidi, Resolve AI
Creator perspective · MoneyVarun argues context windows are not the only problem: with large enough context, models over-explore, generating hypotheses that don't exist or won't lead to the right answer; with very limited context, they under-explore and miss pathways. For production incidents, context becomes a huge issue because telemetry is literally infinite — you can create as many log lines and metrics as you want — and when infinite telemetry is married with code and infrastructure, real cracks in the system start to appear.
AI Engineer · The 6 Pillars of an Agentic Harness for Production — Varun Krovvidi, Resolve AI
Creator perspective · MoneyVarun argues that associating a single model to a workflow or outcome is an overlooked mistake: every workflow has hundreds of different types of tasks — reasoning, deterministic, image reasoning, reasoning across images, SQL, or logs — and each frontier model is good at a specific kind of task. The orchestration engine needs to keep pace with how models evolve and marry the best model to the best task.
AI Engineer · The 6 Pillars of an Agentic Harness for Production — Varun Krovvidi, Resolve AI
Principle · BodyFor absolute strength movements like back squat, front squat, bench press, and deadlift, measure average (mean) velocity, which captures the speed of the entire rep including eccentric and concentric portions. For technical coordination movements like cleans, snatches, jerks, and push presses, measure peak velocity to capture the highest instantaneous bar speed.
Garage Strength · I Had 10 Athletes STOP Lifting Heavy. They Got STRONGER.
Creator perspectiveFor an original film like TAR, marketing money must go to making people care about something they've never heard of — an 'education tax' that remakes avoid entirely.
Slidebean · Why Hollywood Keeps Making $125M Flops
Creator perspective · MoneyA POC built on around 40 pre-curated files can validate parsing, chunking, OCR, embedding, vector DB, and knowledge graph, but when scaling to 80,000+ files the bottleneck is file discovery, sensitivity scanning, and data quality — not the AI model itself. Plan for scale from the start.
AI Engineer · From Raw Documents to AI-Ready Data — Leo Platzer & Jeff Koss