Rivian needed a new factory for its cheaper SUV, so it built the car in the old one and pocketed $2.25 billion. Its CFO's rule is to wring every last drop out of what you already own before you pour concrete, the hardware version of staffing for the work you have, not the work you hope to get. The new model pulled profitability forward instead of pushing it out, because sharing the factory spread fixed costs over more cars and made the older lines cheaper too. Revenue tripled in two years while operating costs held flat. Sometimes the cheapest way to grow is to not build the new thing.
Startups in this program have closed deals with customers like Visa, Meta, Unilever, GM, AstraZeneca and Walmart. Build with the coding assistant used by 90% of the Fortune 100. Scale on AI-ready cloud infra backed by enterprise-grade security, privacy, and governance. Sell on Marketplace.
Z.ai's GLM-5.2 flagship model marks a substantial leap in long-horizon capability over GLM-5.1. It has a solid 1M-token context window and new capabilities. Its improved architecture reuses per-token FLOPs by 2.9x. The model has been released under an MIT open-source license.
Self-serve access to ChatGPT Ads is now open to all advertisers in the US. OpenAI crossed $100 million in annualized ad revenue in six weeks with less than 20% of eligible users seeing ads daily. 85% of free and Go tier users are eligible to see ads. OpenAI has confirmed rollout to Canada, Australia, New Zealand, the United Kingdom, Japan, South Korea, Brazil, and Mexico.
Lemkin names the "20-year tax" where loyal customers get squeezed because they've already signed, integrated, and trained their teams while new logos get war rooms, CEO involvement, and free custom integrations, a pattern running through nearly every pre-AI B2B company where the CSM rotates every 14 months and nobody remembers the case studies or referrals.
The venture-backed rocketship is just one way to build a company, and you absorbed it without noticing. Patagonia stopped chasing growth, let customers set its pace, and added a mission that cost real money, and its near-death cash scares went away. Warren Buffett took the cash off a dying textile mill's books, bought stocks with it, and grew Berkshire into a giant without raising a dollar of venture money. "Raise, grow fast, exit" is a script, not a law, and there are far more ways to win than the one everyone copies. The startup is one configuration. Is it the right one for the life you want?
The Rule of 40, useful in SaaS for balancing growth and profitability, often misrepresents hardware companies due to longer development cycles and upfront costs.
The sequence of mountains is when founders fixate on the mountain in front of them, putting everything they have into climbing it, only to realize there are other mountains behind it.
The list is in your head. The messaging too. What's missing is the four hours it takes to find the right contacts, write sequences that don't sound generic, and actually send them. Every single week.
The AI industry should be building a frontier ecosystem so that value flows broadly across every company, industry, and country. This ecosystem would own the learning loop that encodes its institutional knowledge, compounding its human and token capital. Platforms enable more value on top than is captured inside. Once implemented, companies will be able to create value for themselves and for the economy around them.
Most people who try AI are just occasional users. Large chunks of the population aren't using AI at all. AI use hasn't shifted much in the past six months to a year. The only thing that has substantially changed in that time is that the negative sentiment about AI has gone significantly up. Many people are holding back on AI use because of real AI concerns and a lack of perceived AI value.
AI investing breaks software's pattern-matching because the system has more variables, they couple unevenly, and each decomposes into sub-variables on independent curves, so solving one constraint just changes which constraint matters next.
AI-forward companies are now reaching $10M to $25M in revenue with about 20 go-to-market people, while the rest of the field still needs closer to 35 to achieve the same result. The benchmarks under that shifted too, with median net revenue retention sitting around 108% to 110%, the top quartile holding above 123%, and more comp plans tying pay to net-new recurring revenue. Pricing is the other big move, as outcome-based and usage-based models spread because they track the value you actually deliver and can pull gross margins toward 90%.
Everyone building AI sales agents reaches for voice first, but the data keeps showing that around 85% of prospects would rather handle it over chat, which is also far easier to deploy, monitor, and fix when it breaks. The teams seeing real results point AI at the routine top of the funnel, qualifying and triaging, then hand the high-value conversation to a human the moment nuance matters. If you are wiring up an AI sales motion, start where it actually converts, and resist making the machine do the part people still want a person for.
Customer.io is the customer engagement platform for tech-savvy marketers. Trigger multi-channel messaging powered by real-time data. Build journeys, run experiments, and iterate without engineering. Join thousands of founders using Customer.io for onboarding, activation, and lifecycle messaging.
The Fusion API allows OpenRouter users to achieve Fable-level intelligence at half the price. Panels of models consistently outperform individual models. Beyond-frontier model performance can be achieved with frontier panels at much lower cost. Fusion creates a panel of several models, each with web search and bash tools enabled, to increase performance. A judge model reads every response and extracts the structure, then a synthesizer writes a final answer grounded in analysis.
The Open Knowledge Format (OKF) is an open specification that formalizes the LLM-wiki pattern into a portable, interoperable format. It is a vendor-neutral, agent- and human-friendly standard for representing metadata, context, and curated knowledge. The specification formalizes the small set of conventions needed to make LLM wiki patterns interoperable. This post details what OKF can solve for organizations and how to get started with it.
For decades, you picked a side, a company sold to businesses or to people, and the whole playbook followed from that one choice. The fastest-growing AI-native companies have stopped fitting in either box and increasingly win both at once, which scrambles how you price, sell, and defend the business. The moats that hold up are the unglamorous ones - proprietary data, the workflow loops people live in, network effects, and brand - the trust that makes someone pick you once the underlying model is a commodity.
Your forever-home on the web should be easy for agents to find and inviting for humans to interact with. Marketplaces and social media profiles that require a login don't fit the bill - your website still matters!
Demo days are packed and $500M funds keep getting announced, but the cash actually flowing back to investors has slowed to a trickle, because the IPO and acquisition exits have been welded shut for years. There are now more than 1,500 unicorns worth around $6 trillion with nowhere to go, and at the current pace, one estimate says it would take 49 years to clear the backlog. Look at the secondary market, where stakes change hands at 40 to 60% below their official marks, the real write-down happening where the quarterly letters won't show it.
CSM hiring flatlined for four years while Forward Deployed Engineering grew 1,000%+, and Assembly AI's Ryan Seams confirmed it filled a TAM pipeline overnight by switching the title to FDE, Lovable's Monica Perez rebuilt CS on Lovable replacing Gainsight and Harvey's Tom Ronen stayed traditional with on-site EBRs because selling AI to law partners is change management not software. The panel killed NPS unanimously, agreed 50%+ of CSM activities have no correlation with retention, and Bobby Cooper's Weave case study showed moving closed-won to an implementation success threshold dropped churn from 4% to ~0.5% monthly while scaling $8M to $200M ARR.
Most AI bills balloon for a dumb reason: companies run frontier models for everything, when open-source models are now nearly as good and often 90% cheaper. One team swapped off Opus and cut costs by 93% with nobody noticing. The fastest win is routing: send only the genuinely hard work to the expensive model and push the rest to a cheap one, and the bill drops the same day. The other money pit is rewarding raw usage, because the minute you post a most-tokens-used leaderboard people burn cash to top it, so track spend per person and judge everyone on impact instead.
When a market is changing fast, buyers stall because they're scared of picking something they'll regret in five years, and what reassures them is hearing your clear view of where the whole market is going. This is your belief about where things are heading, the assumption your roadmap rests on, and it should be rooted in what you do better than anyone. Notice how the giants each push a future that conveniently centers their own strength: OpenAI says AI is a metered utility, Microsoft says models are commodities and context is king, and ServiceNow says it's all workflows and governance. Stake out yours just as deliberately.
Stripe is expanding Stripe Projects with new agent integrations, additional providers, and new developer controls. Stripe Projects is now available as a skill in Hermes, an open source self-improving AI agent. Stripe Projects handles infrastructure provisioning and management, while Hermes carries context across sessions. The update gives developers working on complex products over days or weeks access to a persistent AI collaborator that returns to a project the same way a human developer would.
Google is introducing new Gemini app features specifically designed for business owners that transform the AI into a deeply knowledgeable, in-pocket partner that natively understands the business and helps business owners get more done. The new features will connect directly to users' Google Business Profiles to access real-world context like customer reviews, customer questions, and performance data. They will allow business owners to offload complex, time-consuming tasks, like turning insights into action. The updates are rolling out globally this month.
AI will transform white-collar work, but the real value lies in how companies harness human creativity and judgment that AI cannot replicate. Companies leading in this transformation are integrating AI to enhance internal knowledge and decision-making, turning individual insights into institutional advantages.
Anthropic's Fable AI imposes crucial guardrails while offering unprecedented power, evident in Stripe's rapid Ruby codebase migration and enhanced AI benchmarking scores. This setup ensures stability in tech and finance amidst evolving AI capabilities. The AI landscape shifts rapidly, necessitating adaptive strategies for maximizing these powerful tools safely.
Your opinion on AI is only worth something if you earned it by actually using the tools, rather than absorbing a position by proximity and defending it like a team you root for. Ask yourself what would change your mind, because if your honest answer is nothing, you've stopped holding a view and started defending an identity, whichever side you're on. Being early counts for less than staying clear-headed and updating as the evidence changes.
2026.06.12·
theengineeringmanager.substack.com·8 min read·TLDR Founders
Many people struggle to understand probabilities, leading to stress and misaligned expectations, especially in areas like college admissions and investments.