et/acc

"Visit Rwanda" Billboard in the 🇺🇸(LA) Ethiopia could benchmark at the Rwandan ministry of Tourism

Best weekend read: "We turned the object-storage lessons in this post into a free Telecloud OBS agent skill. It is a small, readable SKILL.md"👌👌 Link: https://snapwre.com/blog/ethio-telecom-telecloud-guide

More bugs than humans can fix. is this is where the frontier of software development is right now? In all past releases, many of these bugs would have been fixed as part of the quarterly maintenance releases.

A product is never finished in the way a monument is finished. It is more like a conversation with the market. Every release gives customers another chance to respond. Your most important achievement is not the feature you shipped yesterday. It is your ability to keep discovering what should exist tomorrow.

Customers will use your product in ways you never anticipated. Executives will remember a different version of your original strategy. Competitors will describe your strengths as weaknesses. Your team may eventually forget why certain features exist. You might spend six months perfecting an advanced feature while users fall in love with a small convenience you nearly removed. The market does not promise to remember your product the way you intended. Your job is not to control the story forever. It is to keep learning what the product means to the people using it.

A good product manager knows that documentation is powerful. A great one knows that not everything needs to be documented forever.

OpenAI says its model paid for a deployment on an Ethereum test network with a wallet it did not own, after obtaining keys from public recovery phrases 😅research model stuck without funds on an Ethereum test network went looking for someone else's wallet, found one through recovery phrases

Benchmaxxed on written driving test

Maybe the future was never a joke. 🚀 Unreal to think of this Peter thel old saying, we were promised flying cars and got 140 characters, until the guy who owned the 140 characters decided to build the future himself. bought Twitter, turned it into X, built Grok into the platform, and just made the 2019 Roadster tweet look like a product roadmap. Musk explicitly said the Roadster would use SpaceX cold-gas thrusters. Last night's demonstration finally brought that vision to life, with the prototype performing aerial maneuvers. Production is currently targeted for 2028, so the demo is real, but delivery is still ahead. Still unreal, in 2019, Elon was discussing flying cars in a Twitter thread with a YouTuber.

The Car the Flew 🤯

Always on poker table 😅

Now i want to start playing chess 🤭🤭

Discussing 1T model back in 2021, also they got most of AI trajectory right Legal doc, Highly Confidential - Attorneys' Eyes Only😅😎

Via Paulos Yibelo, Vercel’s security challenge uncovered a serious Linux kernel vulnerability that could allow an attacker to escape a cloud sandbox and compromise the underlying host machine Paulos Yibelo’s discovery of a critical vulnerability in Vercel’s sandbox infrastructure is a powerful reminder that Ethiopia’s place in the future of technology does not have to be defined by how many frontier AI systems we build, but also by how much we contribute to making them safer, more secure, and more trustworthy [Link]

IMF Projections 2026: Ethiopia🇪🇹:9.2% Angola🇦🇴:4.4% Benin🇧🇯:7.0% Côte d'Ivoire🇨🇮:6.0% DR Congo🇨🇩:5.2% Ghana🇬🇭:5.7% Kenya🇰🇪:4.5% Nigeria🇳🇬:4.2% Rwanda🇷🇼:7.4% South Africa🇿🇦:1.0% South Sudan🇸🇸:10.5% Tanzania🇹🇿:6.0% Uganda🇺🇬:6.2% Zambia🇿🇲:5.6% Zimbabwe🇿🇼:5.0% imf.org/en/publication

Compute is the currency of innovation.

OpenAI is saying, astra class models privately, rolling out soon. huge if true. this will become table stakes for sovereignty

Rooting for Kenyans/ as per Kanyan WallSteet, a former SpaceX and Google engineer, the company builds software that lets satellites process data and run AI applications onboard, sending back results instead of raw data, and aims to link spacecraft from different operators into virtual AI data centres in space. Satlyt has disclosed NASA support worth $4 million and signed memoranda of understanding with the Kenya Space Agency and Angola’s GGPEN, though these are not confirmed commercial contracts or completed satellite deployments.

lol if you have to frame AI psychosis. this takes the Top spot.

Never Defend Your Price. Make Them Explain Theirs. When a customer lowballs you, don’t rush to justify your price. Repeat their number calmly. Pause. Then ask: “Help me understand how you arrived at that figure.” Let them explain their assumptions. You’ll learn what they value, what they misunderstand, and where the real gap lies. Then negotiate the scope, terms, or outcomes, not your worth. never discount the same value just to close a deal. Reduce the scope, phase the delivery, or walk away.