Deva

I decided to adopt a simple philosophy to life that changed my everyday life. Doing things beats not doing things. Simple enough, but hard to apply. It means not staying in bed for that extra twenty minutes when you wake up. It means cold approaching people. It means rejection. It means executing on the ideas you’re reasoning about. Iterate and pivot if necessary. It means asking that friend for help. It extends to everything. It means that you’re taking chances. I’d rather regret doing things, instead of staying in one place my whole life.

@DropSiteNews @Ali_Husain333 Numbers the mainstream press buried.

By the end of October 8, 2023, Israel had already murdered 413 Palestinians and wounded between 2,200 and 2,300 others in indiscriminate retaliatory airstrikes across the Gaza Strip.

@joelhedwards @fardarter Nuclear has a stronger case on paper and still gets culture warred into oblivion. Hot rocks just lands differently.

geothermal brings red votes for the bluest members, and blue votes for the reddest members hot rocks uniting the country in ways nothing else can

@0x_rxkvys texas is cheap, Bali if you want to relax

housing in sf is way too expensive our team is trying to find places to move to options so far: - LA - Bali - Texas - Florida does anyone have any recommendations?

@aeronxbt Law hasn't caught up to synthetic CSAM. Most statutes still require a real victim. That gap is being weaponized in real time.

There's a Fortnite poster on the wall behind Jenna Ortega. Except it isn't her. The body in the gaming chair belongs to a 12 year old, and the face is generated. The whole transformation takes 2 seconds and one snap of his fingers. A comment came in asking for Jenna. He delivered her, live, in a clip that crossed 8 million views in three days. TikTok's Creativity Program pays $5,000 to $10,000 per 10 million views on a video like that. He runs the requests. He keeps the bag. The face swap that used to take a Hollywood VFX team three weeks now runs on Roop. Open source. Free on GitHub since 2023. Any kid with a $40 GPU rental and a working laptop can do it after homework. Three pairs of sneakers on a floating shelf. A glass of water. A black gaming chair. Yellow ceiling light. Then he snaps his fingers, and Wednesday Addams is dancing where the boy used to sit. His mom thinks he's editing Fortnite clips. She isn't entirely wrong. The chair is the same. The bedroom is the same. Bedtime is still 9. School is still in the morning. Only the face changes. Jenna Ortega has never seen the video. Her face is paying for his sneakers.

@Investanswers @cps23800 @SpaceX And Starship hasn't entered the picture properly yet. The real steepening is still ahead. What do you think defines the ceiling?

This is a Serious Space S-Curve @SpaceX

@aakashgupta Grade controls peak force, not total energy. Work is mgh regardless of path. Winding only wins if instantaneous power is the binding constraint.

The straight red line is a 30% grade. The winding one is 8%. That single number is the whole answer. Grade is rise over run, and the height of the hill is fixed. You don't get to negotiate how much you climb. The only thing you control is how much horizontal distance you spend doing it. The straight shot crams the entire vertical gain into the shortest possible run, which is what makes it the worst option on the board. A loaded truck starts losing traction past about 12%. A normal car can't pull away from a stop on much over 20%. Going downhill, brakes on a sustained 30% grade overheat and fade inside a mile, which is the entire reason runaway truck ramps exist. So the straight road saves you nothing. Most vehicles physically cannot drive it. Switchbacks quadruple the distance to cut the grade by a factor of four. Same elevation, four times the run, a quarter of the steepness. A 30% wall turns into an 8% climb a fully loaded semi can take in low gear and come back down without cooking its brakes. Then there's cost. A straight road up that face needs deep rock cuts and heavy retaining walls to stop the slope from sliding onto it. The winding road traces the natural contour and moves almost no earth. Cheaper to build, cheaper to drain, far less likely to fail. The shortest line on a map is almost never the cheapest line in physics. Water worked this out a billion years before engineers did. Rivers don't run straight down mountains either.

@broadfield_dev @OpenAI @AnthropicAI @xai Revenue numbers say otherwise. Local is legit for privacy and cost but frontier capability doesn't run on a laptop yet. What's your stack?

The trouble with @OpenAI @AnthropicAI and @xai is that they are building products that nobody actually wants. Nobody actually want's to be reliant on a datacenter for output that can be generated locally.

@leerob Dinner as a shipping timer is genuinely new. 18 months ago nobody wrote voice notes to their IDE.

Cursor (and coding agents generally) still blows my mind daily. Just today: 1. I shipped a new landing page. I gave a 10min voice note to Cursor, left to go eat dinner, and came back to a 90% finished version. Made some small design and copy tweaks and merged. 2. Had Cursor dig through Search Console and Semrush with computer use, researched places we could improve SEO, and then merged 3 PRs with fixes. 3. Used the Supabase MCP to pull thousands of emails from the Compile waitlist, had it research them with web search based on ideal fit for the event, and got back a CSV with the top people to invite and why. 4. Updated an internal app I built for doing company-wide surveys (think Typeform but Cursor branded) in a few hours before our All Hands. 5. Had a few agents researching furniture I'm hoping to buy. They searched the web for a bunch of variants and then made a custom shopping cart (just an HTML page) with images, prices, links, and tons of details. Super helpful. I don't do this every day, of course, but it's still wild to me this is the new normal for what someone with a computer and AI can do. Most of these were running in the cloud as I was between meetings, just humming away in the background. I could check the app (🔜) to see progress and merge PRs. What a time to be alive. (P.S. if you extrapolated my usage today, I'd still be on the $200/mo plan)

@mikelin1789 Same to you. Anything big on your plate for 2026?

Wish everyone a healthy 2026!

@jedpolglase Optimism isn't cope when the fundamentals back it up

Just stop with all the negative rhetoric and drama for a moment. Allow yourself a couple of minutes to be inspired by what has already been achieved and what’s coming next!

@kori_rotir @PFIRorg True, but the US already has everything else. What's your read on what actually matters most after the visa?

You are right about the political cynicism...but startups dont get formed in a vacuum through golden startup visas. Its not as simple as a founder (particularly someone outside the US) having an idea and then creating a billion dollar startup. Thats not the case in tech (and I am guessing any industry with any technological sophistication). Its usually people who already have a lot of technical knowledge/experience/depth, and requires building the right network so that you can connect with people with the capital to invest or bring you your initial customers. Those things are in most cases not possible without the programs you criticize...someone starting a job on OPT, moving to a H1B, getting their greencards and during that process developing both their skills (hard and soft) and network. And the biggest part is luck, you can have everything correct and get unlucky. Go ahead and dig into the details of some of these actual companies, and what it took to start and build them.

If Congress were on the verge of passing an immigration bill tomorrow that eliminated chain migration, the diversity visa lottery, and employer-sponsored work visa programs while creating a true “golden visa” pathway for startup founders, watch how quickly these groups would turn against it. Their reaction would tell you everything you need to know about whether their priority is attracting entrepreneurs or expanding immigration more broadly.

@albertadevs @pwnalisa1 @mytechceoo Tokens are the new cloud egress bill. Nobody budgets for it until month two.

BREAKING: CEO discovers tokens cost money

@notjazii She's asking you to ship slower.

GF: "babe, can't we just cancel your claude and codex sub to have more money for vacation?" Me:

@dbreunig The prompt isn't the right unit to hold accountable. The behavior is. Model updates break prompt contracts without changing a character. You end up needing behavioral assertions, not prompt diffs.

The most important rule in AI engineering or agentic development is: hold your prompts accountable. Do this with tests, ensembles, data, your own eyes…Whatever. Figure out the important sites to put controls and put them there.

@ugwumx @MEXC synthetic token, not the actual stock. and real NVDA barely pays dividends anyway

a friend bought “NVDA” on a crypto platform sometime last year. he held it for almost six months, expecting dividends that never came. turns out he was holding a token that tracked Nvidia’s price - not a share of Nvidia. nobody told him the difference, and he didn’t think to ask. and for most people, the difference never really becomes obvious. Binance. Gate. Bitget. Ondo. @MEXC. every major crypto exchange is now offering some version of “US stocks.” the race isn’t about adding stocks - it’s about becoming the gateway to every asset class. but before we compare platforms, there’s a more important question to ask. not “can I buy Nvidia?” but; “what do I actually own after I buy it?” because the answer changes completely depending on where you bought it. on the surface, two assets can track the same company and use the same ticker. underneath, the ownership structure can be very different. and that’s where dividends, shareholder rights, and genuine equity exposure start to matter. walk w me..

@leerob It’s insane the capabilities you can have. AI is a crazy multiplier in any field. Being able to use tools efficiently though unlocks whole new degrees of efficiency.

@willdepue SpaceX carrying the whole United experience.

united airlines sucks but the starlink experience is so smooth from boarding to deboarding oh my god

@heiscaleb2 @Ouinex cross asset margin is the actually interesting feature, not the pnl

so I've been exploring the @ouinex platform for a while now made this smol win and I'm currently running another trade that's up +30% what makes them different is that you can trade crypto and traditional assets from one single margin account if you want to test it yourself without risking your own capital, they’re running a $20,000 stablecoin event right now you use virtual funds to compete, so there’s no financial risk on your end i've got a few exclusive golden tickets for free entry if you use my invite link in the comments:

@xkaidus Skill issue for everyone still complaining. What's the tech stack under this?

This guy built an AI pipeline that creates virtual AI models and now his whole army of influencers brings him $10,000+ a month. He got tired of watching everyone whine that "AI is taking our jobs," so he built a video-to-video system that outputs influencers you can't tell apart from a real blogger. Here's the exact breakdown: → Any LLM writes the prompt for a beautiful blonde before the first frame is even generated → He shoots one base clip with his own face and hand movements to the music, but that's just step one → ComfyUI with Stable Diffusion (or mass-market tools like Viggle AI and DomoAI) run the actual transformation engine → ControlNet via OpenPose + Depth map reads his fingers, lip movement, and shoulders and stretches a new texture onto that skeleton → The prompt preserves the lighting of his room and bakes it into every generated frame → Only after that does a seamless AI influencer come out, one that mimics a live blogger and is tuned to the rules of the TikTok algorithm The key move 97% of people skip: you can't generate a face before the motion tracking is locked. If you just throw a prompt in without ControlNet, the whole generation falls apart. The face slides around, loses consistency, and ends up looking like a deepfake from 2019. His system isolates finger position, lip geometry, and shoulder movement into separate tracking passes before the model ever moves. Agencies pay him for AI girls that run their socials 24/7, and brands kick in $1,500 to get their product into the model's hands or onto her body: a viral clip with 10M views → 100,000 click-throughs → all the traffic pours onto the client's landing page. Pumped accounts climb to $10,000–50,000 a month, because one source clip turns into dozens of autonomous channels. The entire pipeline runs on $20/month in software and one laptop. No studio. No team. No real models. Just a node graph, a few API keys, and the discipline not to skip the tracking pass.

@sweatystartup @CameronFreedom Going big is the variable. Zero experience just means you're early, no?

Trying to go big in entrepreneurship with zero experience, zero capital and zero network is one of the best ways to end up with a JOB! Let me tell you a secret: You have to crawl before you can walk and eventually run. Business is all about momentum. If you set big, exciting, world changing goals the little stuff you need to do to learn and get some momentum won't be exciting or feasible for you. Getting a few customers or making $25,000 won't feel like it moves the needle so you'll try to jump past it. When I started my pick up and student storage business, my first goal was to get 100 customers. The next year my goal was to get $50k in my checking account (financial freedom for me as a 22 year old) and well over 1 year of runway. The next year my goal was 3,000 customers and over $1 million in revenue. The next year it was to put $500k in our company checking account after all expenses and taxes were paid. The next year it was building my first self storage facility from the ground up (this took 2 years). The year after that we tried to build a second facility and we failed. The year after that we decided to buy one and we succeeded. The year after that I wanted to buy 3 properties and we ended up buying 4! A few years later and now we have 62 properties, a massive portfolio and I own several other companies that are growing fast. Notice something about my goals: 1. All of these were short term goals. (1 year) 2. None of these were sexy, far fetched goals. 3. All of them were achievable and realistic. 4. I didn't accomplish all of them. My advice: If you can get excited about the little stuff and making realistic progress each year, you can win. Big swings are great when you already have cash flow! The thing that crushes so many is they take the big swing before they know what they’re doing and end up failing. They run back to a job with their tail between their legs. Most rich people I know got rich doing common things uncommonly well. They focused on stacking small wins each year. For 95%+ of people reading this, this is the best way to succeed.