Creator perspective · MoneyGary Vaynerchuk argues that social media is the greatest equalizer the world has ever seen: there is no gatekeeper, and the algorithm doesn't care about status or gatekeeping — anyone can post and reach an audience.
GaryVee · How I'd Grow on Social Media If I Started Today
Creator perspective · MoneyRob Walling argues that a lower price can help gain traction only if you have a real competitive advantage; if price is all you offer, you'll face higher maintenance costs and higher customer churn, become a commodity fighting larger competitors with stronger brands, and struggle to win customers without true differentiation. You need to give someone a reason to go with a no-name over a brand name.
Startups For the Rest of Us · Episode 854 | How to Find Startup Ideas (Sneak Peek Chapter from My New Book)
Creator perspective · MoneyRob Walling explains the biggest drawback of the ecosystem approach: by the time growth is obvious there are usually many competitors, and it's easy to launch a year too early and run out of money before the ecosystem matures. Being early means gambling on market growth, something you can't control.
Startups For the Rest of Us · Episode 854 | How to Find Startup Ideas (Sneak Peek Chapter from My New Book)
Principle · MoneyRob Walling discourages investing time to build an audience just to launch a SaaS product; audiences are better suited to selling courses and other information products, and are rarely the highest leverage path for SaaS.
Startups For the Rest of Us · Episode 854 | How to Find Startup Ideas (Sneak Peek Chapter from My New Book)
Principle · MoneyRob Walling advises that if you have a strong network in a specific market, you should limit your product ideas to that market to gain early product input and access to other people's audiences. He cites Xero building web-based bookkeeping software in the accounting space while QuickBooks was still transitioning from desktop, and Diiforma bringing biotech M&A and venture funding data into an accessible price range.
Startups For the Rest of Us · Episode 854 | How to Find Startup Ideas (Sneak Peek Chapter from My New Book)
Principle · MoneyWhen a founder pitches a startup idea, Rob Walling advises asking what problem it solves rather than evaluating the idea itself.
Startups For the Rest of Us · Episode 854 | How to Find Startup Ideas (Sneak Peek Chapter from My New Book)
Principle · MoneyYou don't want the same model checking the thing that was done — spawn new sessions and bounce them off a contract to get models out of the self-confidence loop; the speaker endorses this as a better pattern.
AI Engineer · Stop Telling Agents How — Dan Barrett, Raymond Weitekamp & Jose Montes de Oca, OpenProse
Principle · MoneyThe guest argues that everyone in the marketing industry will do dramatically better if they understand that the creative — the art — is the variable of success, and that the math is the commodity which is about to become way more commoditized.
GaryVee · Everything You Know About Social Media Is Changing
Principle · MoneyThe guest states that the biggest opportunity is understanding where the attention is that is underpriced and actually consumed — not potentially consumed — and considers the biggest opportunity in marketing to be one where almost nobody else is paying attention.
GaryVee · Everything You Know About Social Media Is Changing
Principle · MoneyDon't be afraid of hiring salespeople — as long as you have lead flow, salespeople always make you more than the price for them. Hiring salespeople is growth spend and very low risk when lead flow exists.
Alex Hormozi · Can I Save This Failing Theme Park in 90 Days?
Principle · MoneyAl Chen's takeaway: the backup/restore status says nothing about your data at that point in time — a bad disk, bad node, bad migration, or replication lag can all cause data to become stale, different from when the job ran. It's about data correctness, not just checking a box to see if the backup and restore job ran.
endorsed
AI Engineer · Shipping AI Into 5 Major Banks Without a 12-Month Compliance Wait — Al Chen, Cisco
Creator perspective · MoneyAl Chen explains that deploying agentic AI within banks takes a long time, and attributes the slowness to the keyword 'production': the use cases that matter are the production applications millions of customers use daily, like the bank's website and mobile app — not prototypes.
described
AI Engineer · Shipping AI Into 5 Major Banks Without a 12-Month Compliance Wait — Al Chen, Cisco
Principle · MoneyFollow a systematic eval process so you can show evidence and validate that agent outputs follow the guardrails and controls in place; use a consistent method to evaluate and measure agent outputs. Putting hallucinated or inaccurate information in front of customers carries liability, financial, and reputational implications.
AI Engineer · Evals for High-Stakes AI Agents — Doneyli De Jesus & Annabell Schäfer, Langfuse
Creator perspectivePamela Fox observes that even with forced source attribution, a model may still provide the ungrounded information while correctly labeling it 'no source provided' — grounding is not fully fixed by prompting alone. She also notes that when a model gives an ungrounded RAG answer, you can iterate on the prompt to force source-grounding rather than abandoning the model, and do a basic search first before sending the RAG prompt to the model.
AI Engineer · The Model Swap Workshop: GPT-5.5 vs Kimi vs DeepSeek vs Mistral — Pamela Fox, Microsoft
Creator perspectivePamela Fox explains that with DSPy you have to rewrite your whole code because your prompt isn't in your code — your constraints are — but this gives you more model independence and flexibility to use more models, especially with smaller models, meaning you can use the cheapest model and still get the results you need.
AI Engineer · The Model Swap Workshop: GPT-5.5 vs Kimi vs DeepSeek vs Mistral — Pamela Fox, Microsoft
Principle · MoneyWhen you need the latest information or information outside your knowledge window, use a search provider rather than asking ChatGPT or OpenAI to answer from its weights.
Also discussed by Richard Socher on Latent.Space (unclassified source — not counted until classified). Not counted as independent sources.
AI Engineer · Backtest Your Agent Against Yesterday's Web — Ryan Ahern, Exa
Principle · MoneyWhen choosing a search provider, evaluate quality, cost, and latency together. When evaluating cost, account for input token costs in your context window, not just the search API cost — search results are tokens that you pay input costs on.
AI Engineer · Backtest Your Agent Against Yesterday's Web — Ryan Ahern, Exa
Principle · MoneyWhen evaluating a search provider, build your own regenerative benchmark against your specific use case. Static published benchmarks get trained on by search providers and labs, who climb the hill and render the benchmark useless.
AI Engineer · Backtest Your Agent Against Yesterday's Web — Ryan Ahern, Exa
Principle · MoneyTrust is essential for agent-to-agent transactions: you need to be sure a service will deliver what it promises at the stated price; without trust no one transacts. Historically, people built trust by transacting with people they knew, then with people a trusted person recommended (like asking a friend for a barber recommendation), and as the world got bigger, through trusted middlemen — marketplaces and town squares, where selling signaled trustworthiness. Modern economies add reviews and consumer protection agencies. In the agentic world there is no equivalent: it's hard to trust a random URL claiming to offer a perfect weather data API.
AI Engineer · Why AI Agents Still Won't Pay for Anything — Jan Curn, Apify
Creator perspective · MoneyAgents working on longer tasks will eventually ask for approval to spend money, so autonomy on bigger tasks may require giving agents a budget.
AI Engineer · Why AI Agents Still Won't Pay for Anything — Jan Curn, Apify