Anecdote · MoneyJim Clark uses a newsroom analogy: the reporter who gathers stories is not allowed to publish to the website or write the final article — roles are separated. Complicated systems are built by splitting them into little pieces with separate roles. In the analogy, the researcher sandbox has web access but no MCPs that can write to the corporate site and writes research to a local temp directory private to that agent; the fact-checker sandbox gets some MCPs but no network access, can read the research done for it, and can access the fact database; the reporter/publisher sandbox has no network access but gets the Notion MCP or publisher MCP because it only sees filtered research.
AI Engineer · How Many Credentials Should Your AI Agent Have? Zero. — Jim Clark, Docker
Recommendation · FuelJesse says even when lean you can still eat pasta, respecting complex carbohydrates rather than eliminating them, since they are important fuel for performance; as long as he remains active he can eat these things, and carbohydrates will always be his primary energy source. To avoid overeating, he uses a plate division method allocating one third of the plate to complex or starchy carbohydrates, because these foods can easily be overeaten without portion allocation.
ATHLEAN-X™ · 17 Foods That Make it EASY To Get Lean (AND STAY THAT WAY!)
Anecdote · MoneyYugabyte spent almost a year trying to ship four agents and failed to make them work because of the data problem. Their three failed approaches: throwing everything into a knowledge base with a distributed RAG pipeline didn't do anything good; letting every agent spin up its own tiny Postgres database failed because every repeatable workflow was being reinvented on the fly; and a backend with no UI for agents failed because they couldn't tell what was going on — no human in the loop equals really bad results.
AI Engineer · Agent Memory Is Solved. Agent Learning Isn't. — Karthik Ranganathan, Yugabyte
Anecdote · MoneyWhen drowning in inbound volume, AssemblyAI's natural instinct was to deflect more conversations and talk to fewer people. They bought an off-the-shelf support bot, pointed it at their docs, and expected it to answer all docs questions so they'd never have to answer simple customer questions again. It only resolved about 10% of conversations end-to-end, leaving the team to field roughly 900 tickets a day — not good enough. Worse, they had no access to the system prompt, tools, or RAG infrastructure, so they couldn't iterate fast enough; every needed change was deferred to the vendor's roadmap. That didn't work, so they decided to build their own.
AI Engineer · We Built an AI Support Agent That Resolves 80% of Tickets — AssemblyAI
Recommendation · MoneyThe source recommends setting SMART marketing goals — specific, measurable, achievable, relevant and time bound — giving the example of 'increase sales by 10% in the next 12 months' instead of a vague goal to 'increase sales'. It also advises tying marketing goals into the overall goals in the business plan so marketing supports broader business objectives.
Offered as best-practice guidance; the article provides no evidence that SMART goals improve outcomes, only an illustrative example.
Commonwealth of Australia · Develop your marketing plan
Report · MoneyCharlie Holtz's first principle for being a top builder is to stay near the frontier — always trying the latest tools basically the day they come out, such as Ultra Code or slash goal. He argues this matters because the field moves so fast that relying on your social graph for workflow information leaves you three to six months behind. For startup founders, staying near the frontier is also a source of new ideas for what to build; for employees, it means being the person at the company who always knows the latest workflows.
AI Engineer · Orchestras, Not Factories: How the Fastest Builders Work — Charlie Holtz, Conductor
ReportIf a failure occurs before the mutation, retrying is generally safe because nothing has changed; if it occurs during the mutation, the system may be partially modified and its state needs to be inspected; if it occurs after the mutation, the operation may have committed even if the acknowledgement was lost, so retrying could create duplicates.
speaker and stance unattributed in source
AI Engineer · Your Coding Agent Can't Undo Its Own Mistakes - Tanmay Sah & Dolly Sah
ReportIf nothing has changed in the system, the agent should simply retry to achieve the desired state; if the original state is available, the agent can restore it; if an external action has already been committed, a compensation mechanism (compensating action) is needed to reverse its effect; if recovery cannot be verified as safe, or the verifier rejects the recovery, the agent should stop and escalate.
speaker and stance unattributed in source
AI Engineer · Your Coding Agent Can't Undo Its Own Mistakes - Tanmay Sah & Dolly Sah
Creator perspective · MoneyOn the choice between using AI to scale and leaning into authenticity as a creator, Gary Vaynerchuk's take is that everyone should fall in love with 'and', not 'or'. The host adds that one of the best things a human creator can do today is lean into their authenticity and humanness to build their audience, and that AI is really best at creating more content — so a creator can use AI to scale, or deliberately choose not to use AI and potentially use that as a way of standing out.
GaryVee · You Might Be Making Content Completely Wrong
Creator perspective · MoneyNew models are released every couple of weeks, making evals more important every month. Without evals you have no way of knowing which things a new model is better or worse at before switching. With evals, you run the suite, compare the numbers, and know whether you can upgrade your model. The workshop starts with Claude 4.5 and upgrades to Claude Haiku 4.5 (released about a week prior) to demonstrate model comparison via evals.
AI Engineer · How to Know If Your AI Agent Is Any Good: Evals With Phoenix & Claude — Laurie Voss, Arize AI
Creator perspectiveFailing to restrict agent network access via environment configuration leaves API keys exposed to potential theft; restricting it keeps the key 'hidden away from any potential theft'.
AI Engineer · Why One AI Agent Isn’t Enough (And How to Run 1,000) - CJ Avilla & Harrison Stall, Anthropic
Creator perspectivePutting an API key into a markdown file given to Claude is a security problem — 'that's definitely bad. Do not do that.'
AI Engineer · Why One AI Agent Isn’t Enough (And How to Run 1,000) - CJ Avilla & Harrison Stall, Anthropic
Creator perspectiveStoring API keys as plain-text files on the sandbox disk — even hidden files or secret directories like 'myapikeys.md' — always carries a risk of leaking, whether through prompt injection or other means; Anthropic built abstractions so credentials never have to touch the sandbox.
AI Engineer · Why One AI Agent Isn’t Enough (And How to Run 1,000) - CJ Avilla & Harrison Stall, Anthropic
Creator perspectiveGiving an agent unrestricted internet access — with WebFetch/WebSearch enabled by default — without explicit permission risks the agent reaching unintended domains; you don't want any internet access without explicitly allowing it.
AI Engineer · Why One AI Agent Isn’t Enough (And How to Run 1,000) - CJ Avilla & Harrison Stall, Anthropic
Creator perspective · MindBrooks walks through an exercise of eliminating idols one by one — power, pleasure, money — to reveal what remains: honor. He notes that if asked to name your idol at the start, you might not have arrived at the same answer; the elimination process itself reveals it. Aquinas suggested that once the other idols are eliminated, what you've found is what can make you stumble and distract you from love in your life.
presented as Brooks's endorsed method
Andrew Huberman · The Art of True Happiness | Dr. Arthur Brooks
Creator perspective · MindBrooks identifies the four pillars of a satisfying life as faith, family, friends, and work — to love and be loved means spending time on faith (love of the divine), family (people biologically or by adoption related to you), friends (the people you actually choose), and serving the world through the way you earn your daily bread. Only when you know your four pillars can you pivot to what your soul actually hungers for, and that's where real satisfaction comes.
presented as Brooks's endorsed framework
Andrew Huberman · The Art of True Happiness | Dr. Arthur Brooks
ReportFigma designs can be taken into Figma Make to convert them into real website-driven code; when someone questions whether a Figma design can be turned into code, the answer is yes; MCP (Model Context Protocol) can be used to take Figma code over to Codex, Claude, or another external AI agent for building, and for complex animations MCP with AI or another method can hand off the animation.
Figma · Figma for Edu: Prototyping with animations in Figma
Creator perspective · MoneyWhen using Claude or Codex on the same task, people commonly run multiple stateful sessions (one plans, another reviews; one writes code, another reviews) — effectively a multi-agent system where the human acts as a Cisco router moving packets between the two stateful agents.
AI Engineer · Why Your AI Agents Can't Talk to Each Other (Yet) — Vlad Luzin, BAND
Principle · MindLiemandt argues that a chatbot alone does not produce the learning improvements — generating a personalized lesson for every kid does. Lessons must respect prerequisites (don't give a chemistry lesson to someone who doesn't know algebra), verify student motivation and attention before judging effectiveness since disengaged students distort results, and use spaced repetition with later recall testing to verify the lesson actually transferred to long-term memory.
Andrew Huberman · How to Accelerate Learning & Improve Education | Joe Liemandt
Principle · MindLiemandt lays out a dependency chain: chemistry is basically algebra with cool word problems, so an algebra test predicts chemistry performance; algebra is fast manipulation of fractions; fractions come down to memorized multiplication and division tables. Math more than anything — and reading with vocabulary — is layers built on foundations, and the time to learn algebra is not in the middle of a chemistry problem.
the guest's endorsed heuristics
Andrew Huberman · How to Accelerate Learning & Improve Education | Joe Liemandt