正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

I wrote the full Craftax RL environment + PPO in one CUDA file. zero python in the loop. training from scratch on my RTX PRO 6000 > 0.6s: first wood, first crafting table, first dungeon entered > 1.7s: first zombie + skeleton kills, first bow found > 2.9s: first... show more

Elliot Arledge

42,779 subscribers

50,500 次观看 • 10 天前 •via X (Twitter)

科学技术游戏

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

My first Allsvenskan goal was against Djurgården at Fredriksskans on the 10/05/2008. Before I made my first team debut for Kalmar FF, I spent 3months playing for the reserve team while I training with the first team on a daily basis. My first encounter with Nanne Bergstrand before I signed with Kalmar FF, he told me how much he trusted me and how he will love it that I joint the team after laying down the plans for me in the team. He said "you will be the joker of the team this season" I didn't understand what he meant be that but I didn't care about that because I trusted myself enough that I will definitely play more than a joker role if given the chance. So during this 3months playing for the Reverse Team I was not happy at missing games for the first team at all because training with them made me feel like I can play in this team. During one of the games with the reverse team against Malmo FF away from home was when everything thing changed for me. Playing against them is always difficult even their first team. We won the game 3-4 and i scored 4 scored 4 goals against a very good Malmo FF reserve team and that was the last time I played for the reserves that season. The next day there was no recovery for me so I had to training with the first team as usual. Nanne called me before training and told me I will be with part of the first team match squad moving forward. May 10th I scored my first Professional goal after being on the pitch for 10mins and I did it in style. I was so happy to have scored the last goal of the game to give us a comfortable 5-1 win. The joy and tears in my eyes was the reward for my patience and determination to prove myself that I have a place in the first team. I went on to play 21 games for the first team and became the replacement for our Brazilian star Caesar Santin who joined Fc Copenhagen. Be patient with yourself and trust in your abilities to make it happen.

My first Allsvenskan goal was against Djurgården at Fredriksskans on the 10/05/2008. Before I made my first team debut for Kalmar FF, I spent 3months playing for the reserve team while I training with the first team on a daily basis. My first encounter with Nanne Bergstrand before I signed with Kalmar FF, he told me how much he trusted me and how he will love it that I joint the team after laying down the plans for me in the team. He said "you will be the joker of the team this season" I didn't understand what he meant be that but I didn't care about that because I trusted myself enough that I will definitely play more than a joker role if given the chance. So during this 3months playing for the Reverse Team I was not happy at missing games for the first team at all because training with them made me feel like I can play in this team. During one of the games with the reverse team against Malmo FF away from home was when everything thing changed for me. Playing against them is always difficult even their first team. We won the game 3-4 and i scored 4 scored 4 goals against a very good Malmo FF reserve team and that was the last time I played for the reserves that season. The next day there was no recovery for me so I had to training with the first team as usual. Nanne called me before training and told me I will be with part of the first team match squad moving forward. May 10th I scored my first Professional goal after being on the pitch for 10mins and I did it in style. I was so happy to have scored the last goal of the game to give us a comfortable 5-1 win. The joy and tears in my eyes was the reward for my patience and determination to prove myself that I have a place in the first team. I went on to play 21 games for the first team and became the replacement for our Brazilian star Caesar Santin who joined Fc Copenhagen. Be patient with yourself and trust in your abilities to make it happen.

Abiola Dauda

21,891 次观看 • 2 个月前

Being here on these steps feels spiritual to me. I’ve stared at the Salad Days cover since I learned of Minor Threat when I was 14. I think back to that first mixtape Jared Hein gave me in class sophomore year of highschool. Hearing Minor Threat on that tape for the first time opened a new door I didn’t even know existed. It pulled me into a world that felt raw, honest, and completely alive. The Salad Days cover. Those steps. It resonates with all of us who came up in this scene. It is more than an image. It is a feeling, a moment, a shared memory. This is my second time here. The first was over 20 years ago, and I can’t find that photo anywhere. So I had to take this one for myself. Just to reconnect. To bring myself back to the music and to that version of me that first felt it. Walking up to the Dischord Records house really does feel spiritual and like a time portal to when I was a teenager and all I cared about was going to shows and listening to records with my friends. Sitting on these steps, you can feel the history in a way that is hard to explain. One day I will meet Ian MacKaye. That would be a dream.

Being here on these steps feels spiritual to me. I’ve stared at the Salad Days cover since I learned of Minor Threat when I was 14. I think back to that first mixtape Jared Hein gave me in class sophomore year of highschool. Hearing Minor Threat on that tape for the first time opened a new door I didn’t even know existed. It pulled me into a world that felt raw, honest, and completely alive. The Salad Days cover. Those steps. It resonates with all of us who came up in this scene. It is more than an image. It is a feeling, a moment, a shared memory. This is my second time here. The first was over 20 years ago, and I can’t find that photo anywhere. So I had to take this one for myself. Just to reconnect. To bring myself back to the music and to that version of me that first felt it. Walking up to the Dischord Records house really does feel spiritual and like a time portal to when I was a teenager and all I cared about was going to shows and listening to records with my friends. Sitting on these steps, you can feel the history in a way that is hard to explain. One day I will meet Ian MacKaye. That would be a dream.

Steve Aoki

12,612 次观看 • 3 个月前

[Part 2] How I met BTS V hyungnim at the training camp: Turns out we were in the same platoon. 👤I enlisted on December 11th, and at that time, news spread that two BTS members were enlisting. It was all over the news, and there were tons of people. I went in thinking, “How would we even meet among so many people?” After saying goodbye to my parents, everyone moved out to the field, and it was time to sit on the allocated seats. I sat down, and there was one empty seat in front of me. I wondered why it was empty, and then two people walked in, almost like under a spotlight. I thought, “Ah, That must be BTS V hyungnim.” Then, they were dividing us into squads. I thought, “Could we end up in the same platoon?” And we actually did. We ended up in the same first platoon. V hyungnim was in the first platoon, first unit, and I was in the first platoon, third unit. And so that is how my training camp life with V hyung started. - continued in part 3 (

[Part 2] How I met BTS V hyungnim at the training camp: Turns out we were in the same platoon. 👤I enlisted on December 11th, and at that time, news spread that two BTS members were enlisting. It was all over the news, and there were tons of people. I went in thinking, “How would we even meet among so many people?” After saying goodbye to my parents, everyone moved out to the field, and it was time to sit on the allocated seats. I sat down, and there was one empty seat in front of me. I wondered why it was empty, and then two people walked in, almost like under a spotlight. I thought, “Ah, That must be BTS V hyungnim.” Then, they were dividing us into squads. I thought, “Could we end up in the same platoon?” And we actually did. We ended up in the same first platoon. V hyungnim was in the first platoon, first unit, and I was in the first platoon, third unit. And so that is how my training camp life with V hyung started. - continued in part 3 (

Taehyung Naver

47,127 次观看 • 2 个月前

Looking back on awesome but notoriously tough games from the golden era: Karateka (1984), Jordan Mechner's first commercial game, was an instant hit. Considered a cult classic today, it's arguably one of the greatest games ever made. If you made it past that damn bird, I salute you. If you got kicked in the nuts at the very end, I feel your pain... First time I finally made it to the end (after dying to that damn bird like 50 times), I ran up to Princess Mariko still in fighting stance like a total idiot. Instant kick to the nuts. Game over. No save, back to the start. Lesson learned the hard way: drop into running stance before you reach her. Classic Mechner troll ending. Did you get nut-kicked on your first victory too, or did you wisely bow/run like a gentleman?

Looking back on awesome but notoriously tough games from the golden era: Karateka (1984), Jordan Mechner's first commercial game, was an instant hit. Considered a cult classic today, it's arguably one of the greatest games ever made. If you made it past that damn bird, I salute you. If you got kicked in the nuts at the very end, I feel your pain... First time I finally made it to the end (after dying to that damn bird like 50 times), I ran up to Princess Mariko still in fighting stance like a total idiot. Instant kick to the nuts. Game over. No save, back to the start. Lesson learned the hard way: drop into running stance before you reach her. Classic Mechner troll ending. Did you get nut-kicked on your first victory too, or did you wisely bow/run like a gentleman?

exQUIZitely 🕹️

18,685 次观看 • 1 个月前

update: qwen 3.6 27b dense q4 just one shotted octopus invaders game on a single 3090. hermes agent drove the whole thing, ~41 tok/s gen 21gb vram at full 262k context, thinking mode on. one prompt in and the canonical multi-file space shooter benchmark out, the same exact prompt i ran on qwen 3.5 27b dense back in march on the same card. 3.5 needed one external scope bug fix before the game would even load on first play. 3.6 needed nothing. 11 of 11 files written, 2411 lines of code, zero steering interventions, zero external fixes, playable on first load. 16 minutes 41 seconds wall clock from prompt to playable. consumer tier king on a single 3090 is locked tonight, and the silicon underneath my desk did not change between march and now. the open source ecosystem just moved the floor. watch it ship itself, the full 16 minutes 41 seconds sped to 3 minutes 45, no human touched the keyboard between the first prompt and the final frame.

update: qwen 3.6 27b dense q4 just one shotted octopus invaders game on a single 3090. hermes agent drove the whole thing, ~41 tok/s gen 21gb vram at full 262k context, thinking mode on. one prompt in and the canonical multi-file space shooter benchmark out, the same exact prompt i ran on qwen 3.5 27b dense back in march on the same card. 3.5 needed one external scope bug fix before the game would even load on first play. 3.6 needed nothing. 11 of 11 files written, 2411 lines of code, zero steering interventions, zero external fixes, playable on first load. 16 minutes 41 seconds wall clock from prompt to playable. consumer tier king on a single 3090 is locked tonight, and the silicon underneath my desk did not change between march and now. the open source ecosystem just moved the floor. watch it ship itself, the full 16 minutes 41 seconds sped to 3 minutes 45, no human touched the keyboard between the first prompt and the final frame.

Sudo su

123,876 次观看 • 2 个月前

A 16-year-old American student made $7,000 in his first month after publishing his first game on Google Play - and GPT-5.5 built the whole thing for him. He had zero programming knowledge and no $50,000 dev team. He just opened GPT-5.5, described what he wanted to build, and went to play while the AI wrote every line of code for him. GPT-5.5 built a complete 3D game called Hormuz Escape from scratch - obstacle avoidance mechanics, high score system, particle effects, AdMob integration for ads, and a Firebase leaderboard. All from one prompt in a week of work. A Google Play Developer account cost $25 once and forever. 10,000 active players at $0.03 a day through AdMob is already $9,000 a month in passive income. He stopped at $7,000 in the first month and the number keeps growing. 3 billion Android devices in the world and the Play Store is open to anyone with $25 and an idea. AI removed the last barrier - needing to know how to code. $25 to register, One week of work. $7,000 in the first month.

A 16-year-old American student made $7,000 in his first month after publishing his first game on Google Play - and GPT-5.5 built the whole thing for him. He had zero programming knowledge and no $50,000 dev team. He just opened GPT-5.5, described what he wanted to build, and went to play while the AI wrote every line of code for him. GPT-5.5 built a complete 3D game called Hormuz Escape from scratch - obstacle avoidance mechanics, high score system, particle effects, AdMob integration for ads, and a Firebase leaderboard. All from one prompt in a week of work. A Google Play Developer account cost $25 once and forever. 10,000 active players at $0.03 a day through AdMob is already $9,000 a month in passive income. He stopped at $7,000 in the first month and the number keeps growing. 3 billion Android devices in the world and the Play Store is open to anyone with $25 and an idea. AI removed the last barrier - needing to know how to code. $25 to register, One week of work. $7,000 in the first month.

Sprytix

85,583 次观看 • 3 个月前

Introducing Clippo ✨ There is nothing more painful than streaming to 0 viewers. ( i know this ) Everyday Clippo chops my livestreams into viral shorts and automatically posts them to youtube. The results : 3 to 75 live viewers 100K views on youtube 3000+ watch hours ( im monetized ) Doubled my subscriber count ( 1000 to 2000 ) all in 14 days. The best part is clippo can be used completely agentically. When im done with my livestream I just ask my hermes agent to fire up clippo, and the clips are made and scheduled for days. Zero work on my end. I am offering access to 50 people right now, at 50% off, as part of a limited launch. First come, first serve. Once full, Clippo is closed until I can scale compute further.

Introducing Clippo ✨ There is nothing more painful than streaming to 0 viewers. ( i know this ) Everyday Clippo chops my livestreams into viral shorts and automatically posts them to youtube. The results : 3 to 75 live viewers 100K views on youtube 3000+ watch hours ( im monetized ) Doubled my subscriber count ( 1000 to 2000 ) all in 14 days. The best part is clippo can be used completely agentically. When im done with my livestream I just ask my hermes agent to fire up clippo, and the clips are made and scheduled for days. Zero work on my end. I am offering access to 50 people right now, at 50% off, as part of a limited launch. First come, first serve. Once full, Clippo is closed until I can scale compute further.

Max Blade

14,707 次观看 • 2 个月前

Here’s EVERYTHING you need to know about warm up sets: Most people are doing warm up sets ALL wrong and they’re costing themselves from an efficiency/effectiveness standpoint as a result Warm ups are designed to A) guard against injury B) prepare you for your working sets Warm up sets should NEVER be taken in close proximity to failure…most should see you leaving 6, 7, 8+ reps in reserve (the lone exception to this = sets performed for 1-2 reps @ ~90+ of your intended working weight) This is in STARK CONTRAST to working sets which should all end around 0-2 RIR You probably need about ~2 warm up sets before your first exercise or two depending on the specifics of your programming Here is an example of a typical warm up protocol: - a 5-10 minute brisk walk on the treadmill (optional) - 50% of your working weight on your first exercise for ~12 reps (mandatory) - 75% of your working weight on your first exercise for ~6 reps (mandatory) - 90% of your working weight on your first exercise for ~2 reps (optional) After you complete the working sets for your first exercise, you may want to repeat the last 3 steps prior to performing your next exercise IF it is an antagonist exercise (Ex: Lat Pull-Down after Sternal Pecs Machine Press OR Seated Leg Curls after Hack Squats) After that, you probably don’t need any additional warm up sets unless you are just using them as “primers” (sets to make sure your movement pattern / the machine settings / etc. are on point) - If doing a PUSH workout, I’m probably using 2-3 warm up sets before my first exercise only (assuming it’s a multi joint movement) - If doing a PULL workout, I’m probably using 2-3 warm up sets before my first exercise only (assuming it’s a multi joint movement) - If doing an UPPER workout, I’m probably using 2-3 warm up sets before my first exercise and 1-2 warm up sets before my second exercise (assuming they’re antagonist multi joint movements) - If doing an LOWER workout, I’m probably using 2-3 warm up sets before my first exercise and 1-2 warm up sets before my second exercise (assuming they’re antagonist multi joint movements) - If doing a FULL BODY workout, I’m probably using 2-3 warm up sets before my first exercise, 1-2 warm up sets before my second exercise, and may even 1-2 warm up sets before my third and fourth exercises as well (assuming they’re all multi joint movements — antagonist push/pull movements for both upper & lower) (As you can see, the major downside of FULL BODY workouts is often the excess warm up time needed — this is why they work best for beginners much of the time) General rule of thumb: If your first set of an exercise is NOT your best set of that exercise in terms of performance (assuming we equate RIR), you are NOT warming up properly You should see a fall off in performance as sets transpire on a given exercise…this is a feature, not a bug of productive/effective resistance exercise WHEN WARMED UP PROPERLY

Here’s EVERYTHING you need to know about warm up sets: Most people are doing warm up sets ALL wrong and they’re costing themselves from an efficiency/effectiveness standpoint as a result Warm ups are designed to A) guard against injury B) prepare you for your working sets Warm up sets should NEVER be taken in close proximity to failure…most should see you leaving 6, 7, 8+ reps in reserve (the lone exception to this = sets performed for 1-2 reps @ ~90+ of your intended working weight) This is in STARK CONTRAST to working sets which should all end around 0-2 RIR You probably need about ~2 warm up sets before your first exercise or two depending on the specifics of your programming Here is an example of a typical warm up protocol: - a 5-10 minute brisk walk on the treadmill (optional) - 50% of your working weight on your first exercise for ~12 reps (mandatory) - 75% of your working weight on your first exercise for ~6 reps (mandatory) - 90% of your working weight on your first exercise for ~2 reps (optional) After you complete the working sets for your first exercise, you may want to repeat the last 3 steps prior to performing your next exercise IF it is an antagonist exercise (Ex: Lat Pull-Down after Sternal Pecs Machine Press OR Seated Leg Curls after Hack Squats) After that, you probably don’t need any additional warm up sets unless you are just using them as “primers” (sets to make sure your movement pattern / the machine settings / etc. are on point) - If doing a PUSH workout, I’m probably using 2-3 warm up sets before my first exercise only (assuming it’s a multi joint movement) - If doing a PULL workout, I’m probably using 2-3 warm up sets before my first exercise only (assuming it’s a multi joint movement) - If doing an UPPER workout, I’m probably using 2-3 warm up sets before my first exercise and 1-2 warm up sets before my second exercise (assuming they’re antagonist multi joint movements) - If doing an LOWER workout, I’m probably using 2-3 warm up sets before my first exercise and 1-2 warm up sets before my second exercise (assuming they’re antagonist multi joint movements) - If doing a FULL BODY workout, I’m probably using 2-3 warm up sets before my first exercise, 1-2 warm up sets before my second exercise, and may even 1-2 warm up sets before my third and fourth exercises as well (assuming they’re all multi joint movements — antagonist push/pull movements for both upper & lower) (As you can see, the major downside of FULL BODY workouts is often the excess warm up time needed — this is why they work best for beginners much of the time) General rule of thumb: If your first set of an exercise is NOT your best set of that exercise in terms of performance (assuming we equate RIR), you are NOT warming up properly You should see a fall off in performance as sets transpire on a given exercise…this is a feature, not a bug of productive/effective resistance exercise WHEN WARMED UP PROPERLY

Dean Turner

53,653 次观看 • 5 个月前

My 2025 wrap - Got healed from depression - Built a personal brand from 0 - Worked with top brands/companies in the US - Made $30k+ - Made a lot of new friends - Started a business for my mom (funded it) - Furnished the apartment my family and I were living in (it was almost empty at the beginning of the year) - Attended conferences and events - Hosted my first space with over 4k people tuning in - Posted my first YT video that has over 13k views - Earned 7-8 figures consistently monthly - Hired people - Cofounded a design studio - Started paying salaries (in dollars) - Got gifts from brands - Traveled out of the country for the first time - Relocated - Started dressing better - Booked an Airbnb for the first time - Partnered with Cleva - Went house hunting - Saw the possibility of having my own luxury apartment - Made money in a year than my parents ever thought one could in a year - One of the biggest partners in the church - Celebrated my family's birthday for the first time in years - Moved from someone who didn't have a MacBook to someone who owns 2 MacBooks - Built a complete workspace, but have to do it again since I relocated 🙃 - Ate a lot of food I had only seen in movies - Lost a good friend (not died, we just decided to move on from being friends and wish each other well from afar) - Became a dog mum - Started training others - Leveled up my communication skills - Started learning a new language - Started, paused, and resumed reading - Met and talked to people I admired - Showed a lot of people around me what was possible - Made my parents proud A lot can happen in a year, I'm happy my life proved it ✨

Racheal James

46,511 次观看 • 7 个月前

251208 [SPOTLIGHT] RIIZE's Sohee listed "Sticker", "Kick It" & "Favorite" as some of his most listened to songs during his trainee days! He also defended Sticker's honor against fans in the comments 👊 💂‍♂️You were shocked when you heard the exact same whistle from "Favorite"? You were shocked by that? (whistles Favorite) 💂‍♂️When I joined as a trainee...I started listening to NCT's songs for the first time after joining the company. The one that came up the most was "Sticker"... 💂‍♂️From what I remember, "Sticker"... there are so many but "Sticker" and "Kick it" were the first ones I heard. "Sticker", "Kick it", "Favorite" and "Universe" were the ones I heard the most. 💂‍♂️So, after practicing at the company and getting off work, I'd go to the underground parking lot and get in the car. And in the underground parking lot, there's an echo. 💂‍♂️Why "Sticker"? Guys, "Sticker" is a masterpiece. "Sticker" isn't easy. 💂‍♂️So, there was a time when I was really into "Favorite" and I used to whistle the intro and a girl from the training team praised me. From then on, I thought I should show off my whistling a little bit. I should show off my individual skills. Yes. That's when I first realized it. That was the first time I was praised for my whistling. #NCT127 NCT 127

251208 [SPOTLIGHT] RIIZE's Sohee listed "Sticker", "Kick It" & "Favorite" as some of his most listened to songs during his trainee days! He also defended Sticker's honor against fans in the comments 👊 💂‍♂️You were shocked when you heard the exact same whistle from "Favorite"? You were shocked by that? (whistles Favorite) 💂‍♂️When I joined as a trainee...I started listening to NCT's songs for the first time after joining the company. The one that came up the most was "Sticker"... 💂‍♂️From what I remember, "Sticker"... there are so many but "Sticker" and "Kick it" were the first ones I heard. "Sticker", "Kick it", "Favorite" and "Universe" were the ones I heard the most. 💂‍♂️So, after practicing at the company and getting off work, I'd go to the underground parking lot and get in the car. And in the underground parking lot, there's an echo. 💂‍♂️Why "Sticker"? Guys, "Sticker" is a masterpiece. "Sticker" isn't easy. 💂‍♂️So, there was a time when I was really into "Favorite" and I used to whistle the intro and a girl from the training team praised me. From then on, I thought I should show off my whistling a little bit. I should show off my individual skills. Yes. That's when I first realized it. That was the first time I was praised for my whistling. #NCT127 NCT 127

NCT 127 Central

21,811 次观看 • 7 个月前

👤 “Waiting for the hyungs to shower 1 hour each, can Louis shower first” How is it going lately? 🐘 Yes the hyungs still shower first these days 👤 Do they really spend 1 hour inside? All 3 of them? 🌰 No it’s not like that. Ohyul doesn’t take shower that long but he also do other stuff inside 🌰 I once wanted to let Louis shower first but he seem to give up. He just sit in his room and wait for us 🐈‍⬛ Yes! he just sit quitely waiting 🐘 But everytime I ask Ohyul hyung “Can I shower first?” he said “Just shower” (kind of challenge him😭) 🐕 My room has bathroom, why u don’t shower there? 🐘 Wait a sec, what? 🐘 U said I’m not allowed to use it 🐕 When did I say that? 🌰 U can use Woojin’s bathroom 🐘 Okay okay 👤 How long do u spend time for shower? 🐘 For me 20 min 🌰 10 min 🐕 There’s a problem with Louis. When he shower, he doesnt dry so all the waters dropping on the floor as he walks out. Like foreigner style 🌰 He just wear bathrobe #LOUIS #루이 #LNGSHOT #롱샷

👤 “Waiting for the hyungs to shower 1 hour each, can Louis shower first” How is it going lately? 🐘 Yes the hyungs still shower first these days 👤 Do they really spend 1 hour inside? All 3 of them? 🌰 No it’s not like that. Ohyul doesn’t take shower that long but he also do other stuff inside 🌰 I once wanted to let Louis shower first but he seem to give up. He just sit in his room and wait for us 🐈‍⬛ Yes! he just sit quitely waiting 🐘 But everytime I ask Ohyul hyung “Can I shower first?” he said “Just shower” (kind of challenge him😭) 🐕 My room has bathroom, why u don’t shower there? 🐘 Wait a sec, what? 🐘 U said I’m not allowed to use it 🐕 When did I say that? 🌰 U can use Woojin’s bathroom 🐘 Okay okay 👤 How long do u spend time for shower? 🐘 For me 20 min 🌰 10 min 🐕 There’s a problem with Louis. When he shower, he doesnt dry so all the waters dropping on the floor as he walks out. Like foreigner style 🌰 He just wear bathrobe #LOUIS #루이 #LNGSHOT #롱샷

al🗯️

18,332 次观看 • 1 个月前

Checklists are where processes go to be remembered, not done. Mine had 10-12 steps for onboarding a new partner and still cost me 90+ minutes every close, plus one partner who slipped through anyway. I run a creator-led media business. A signed deal triggers invisible work: folder, kickoff doc, tracker entry (really important), reporting schedule, welcome email, intake questions. We built a beautiful checklist in Google Drive. 10-12 steps based on our first meeting. The checklist remembered everything. It did nothing. And the one time we rushed it, a new partner never made it into my tracker, so they never got their first report. Now when a deal closes, we send Viktor one message: run the new-partner setup from our onboarding checklist in Notion. Folder from template, kickoff doc, tracker entry, reporting schedule, welcome email drafted with intake questions, held for my approval. 10-12 minutes later it's done. We review just one email, then send. 90+ minutes of admin became one message. Nothing skipped since, because nothing depends on memory. By the way, writing down your processes was never the hard part. Running them every time, especially on your best days, is. Your checklist can't do that. An AI employee can. Tell me I'm wrong. Hire Viktor for your team. $100 in credits included, no card. Full link in first comment.

Checklists are where processes go to be remembered, not done. Mine had 10-12 steps for onboarding a new partner and still cost me 90+ minutes every close, plus one partner who slipped through anyway. I run a creator-led media business. A signed deal triggers invisible work: folder, kickoff doc, tracker entry (really important), reporting schedule, welcome email, intake questions. We built a beautiful checklist in Google Drive. 10-12 steps based on our first meeting. The checklist remembered everything. It did nothing. And the one time we rushed it, a new partner never made it into my tracker, so they never got their first report. Now when a deal closes, we send Viktor one message: run the new-partner setup from our onboarding checklist in Notion. Folder from template, kickoff doc, tracker entry, reporting schedule, welcome email drafted with intake questions, held for my approval. 10-12 minutes later it's done. We review just one email, then send. 90+ minutes of admin became one message. Nothing skipped since, because nothing depends on memory. By the way, writing down your processes was never the hard part. Running them every time, especially on your best days, is. Your checklist can't do that. An AI employee can. Tell me I'm wrong. Hire Viktor for your team. $100 in credits included, no card. Full link in first comment.

Sufyan Maan, M.Eng

41,983 次观看 • 4 天前

At 20 years old, Keanu Reeves drove his first car from Toronto to Los Angeles to start an acting career. He got out of the car and the first thing his manager said was: “We want to change your name.” They told him Keanu was “too ethnic.” He drove up and down the beach in Santa Monica trying to process it. He considered Chuck Spadina, after the street he grew up on. He tried Templeton. Eventually he took his first and middle initials, Keanu Charles, and started going by K.C. Reeves, which everyone read as Casey. He was credited under that name in a 1986 episode of Disney’s anthology series. It lasted about six months. At auditions, casting directors would call out “Casey Reeves” and he wouldn’t even look up. He just didn’t recognise it as himself. He went back to his agents and told them he couldn’t do it. The whole time, a line from one of the first plays he ever performed kept running through his head. He’d played John Proctor in The Crucible, and the line was: “Because it is my name. Because I can have no other.” He booked Bill and Ted, Point Break, Speed, The Matrix, and John Wick as Keanu.

At 20 years old, Keanu Reeves drove his first car from Toronto to Los Angeles to start an acting career. He got out of the car and the first thing his manager said was: “We want to change your name.” They told him Keanu was “too ethnic.” He drove up and down the beach in Santa Monica trying to process it. He considered Chuck Spadina, after the street he grew up on. He tried Templeton. Eventually he took his first and middle initials, Keanu Charles, and started going by K.C. Reeves, which everyone read as Casey. He was credited under that name in a 1986 episode of Disney’s anthology series. It lasted about six months. At auditions, casting directors would call out “Casey Reeves” and he wouldn’t even look up. He just didn’t recognise it as himself. He went back to his agents and told them he couldn’t do it. The whole time, a line from one of the first plays he ever performed kept running through his head. He’d played John Proctor in The Crucible, and the line was: “Because it is my name. Because I can have no other.” He booked Bill and Ted, Point Break, Speed, The Matrix, and John Wick as Keanu.

Dr. Lemma

2,678,083 次观看 • 4 个月前

This is crazy! With LTX 2 in ComfyUI i was able to keep my original camera movement + it did a full lip-sync on the singing haha. All running locally. I made the original video back in 2021 when Metahumans first came out and we were working in Unreal Engine 4. This music video was some kind of result from my first experiments with Livelink and Control Rig at that time. Now the new AI improved version is not perfect and i didn’t train any lora for character consistency yet. I also didn’t have the original video file anymore so I downloaded the Youtube version as input haha. However the camera result + facial expressions are really promising! I wonder how it would handle more complex and fast shots though.

This is crazy! With LTX 2 in ComfyUI i was able to keep my original camera movement + it did a full lip-sync on the singing haha. All running locally. I made the original video back in 2021 when Metahumans first came out and we were working in Unreal Engine 4. This music video was some kind of result from my first experiments with Livelink and Control Rig at that time. Now the new AI improved version is not perfect and i didn’t train any lora for character consistency yet. I also didn’t have the original video file anymore so I downloaded the Youtube version as input haha. However the camera result + facial expressions are really promising! I wonder how it would handle more complex and fast shots though.

Jerome | InsaneUnreal

41,375 次观看 • 5 个月前

Anthropic Quant Andrej Karpathy to Andrew Ng: "I wrote my own neural net library in JavaScript. Not to ship it - just to see backprop with my own eyes" before any framework, he built convnetjs from scratch - every layer by hand. his advice: skip TensorFlow at first. write it yourself at the lowest level, then pick up the framework once you know what's under the hood. the line that lands: "if all you know is a stack of layers and that dropout helps, you can't debug or improve the model." Watch the talk, then save the loop guide below.

Anthropic Quant Andrej Karpathy to Andrew Ng: "I wrote my own neural net library in JavaScript. Not to ship it - just to see backprop with my own eyes" before any framework, he built convnetjs from scratch - every layer by hand. his advice: skip TensorFlow at first. write it yourself at the lowest level, then pick up the framework once you know what's under the hood. the line that lands: "if all you know is a stack of layers and that dropout helps, you can't debug or improve the model." Watch the talk, then save the loop guide below.

Hanako

61,277 次观看 • 16 天前

$1.7 million in profit flipping land they dont own. Zero money down. With 95% margins. These dudes are Carson and Jackson. They're 23 and 26. Their best month ever was 37 deals and $532,000 in profit. In 30 days. They've never seen most of the properties they flip. They do it all from their phone. This is land wholesaling. But not in the typical way you think. Kind of a reverse method. They aren't lowballing sellers or pretending to be the actual buyer. They start with the buyer first. They call home builders and say: "i see you built nine homes last month. What would you pay for more land?" Builder might say "$50,000 for anything in this neighborhood." Done. Then they find sellers at $40,000 and keep the $10,000 difference. Deal closes in two to three weeks. One guy they know did his first land deal for $110,000 profit in 30 days. In this episode they: - Break down the exact builder-first method they use to lock up deals - Show me the free tools you need to start - Give me the math on how to make $100K a year doing one deal a month - Explain why land has 90% fewer variables than houses - Walk through their $140,000 first big deal at 19 and 21 years old This one blew my mind. So freaking cool.

$1.7 million in profit flipping land they dont own. Zero money down. With 95% margins. These dudes are Carson and Jackson. They're 23 and 26. Their best month ever was 37 deals and $532,000 in profit. In 30 days. They've never seen most of the properties they flip. They do it all from their phone. This is land wholesaling. But not in the typical way you think. Kind of a reverse method. They aren't lowballing sellers or pretending to be the actual buyer. They start with the buyer first. They call home builders and say: "i see you built nine homes last month. What would you pay for more land?" Builder might say "$50,000 for anything in this neighborhood." Done. Then they find sellers at $40,000 and keep the $10,000 difference. Deal closes in two to three weeks. One guy they know did his first land deal for $110,000 profit in 30 days. In this episode they: - Break down the exact builder-first method they use to lock up deals - Show me the free tools you need to start - Give me the math on how to make $100K a year doing one deal a month - Explain why land has 90% fewer variables than houses - Walk through their $140,000 first big deal at 19 and 21 years old This one blew my mind. So freaking cool.

Chris Koerner

61,726 次观看 • 1 个月前

99% of founders are building companies wrong. This is why, after 3 years of development and 14 months securing approval from the European Commission for Educational Content Compliance... I'm pleased to announce my MasterClass: "The Art of Compliance" In this class I will teach you: - Why the cookie consent banner isn't blocking your product—it IS your entire product - Why I rejected Stripe at pre-seed for having ZERO certifications (the Collison brothers personally called me, I hung up) - Why I wake at 3am to physically touch my SOC 2 report before returning to sleep - Why you must hire a Chief Compliance Officer on DAY ONE—before your first engineer, before your first line of code Stay tuned. Access details pending approval from 11 European data protection authorities.

99% of founders are building companies wrong. This is why, after 3 years of development and 14 months securing approval from the European Commission for Educational Content Compliance... I'm pleased to announce my MasterClass: "The Art of Compliance" In this class I will teach you: - Why the cookie consent banner isn't blocking your product—it IS your entire product - Why I rejected Stripe at pre-seed for having ZERO certifications (the Collison brothers personally called me, I hung up) - Why I wake at 3am to physically touch my SOC 2 report before returning to sleep - Why you must hire a Chief Compliance Officer on DAY ONE—before your first engineer, before your first line of code Stay tuned. Access details pending approval from 11 European data protection authorities.

Henrick Johansson

265,250 次观看 • 7 个月前

6 agent patterns for AI engineers: (explained with usage) 1) prompt chaining → split the task into fixed steps, each one checking the last. → use when the task decomposes cleanly and accuracy matters more than latency. 2) routing → classify the input first, then send it to the model or tool built for it. → use when inputs fall into distinct classes that need different handling. 3) parallelization → run several calls at once and merge them, either by splitting the work or voting on the same question. → use when subtasks are independent, or when one answer deserves several opinions. 4) orchestrator-workers → a lead model decides what the subtasks are at runtime, then delegates them. → use when you cannot list the steps in advance. 5) evaluator-optimizer → one model writes, another grades, the loop repeats until it passes. → use when you have clear criteria and iteration measurably helps. 6) autonomous agent → no fixed path. it plans, acts, reads feedback from the environment, and decides when it is done. → use when the steps are unknowable and you can afford the cost and the blast radius. the first five are workflows: you wrote the path. only the last one writes its own. most production systems people call agents are pattern 1, 2, or 5 with good error handling. this taxonomy is from Anthropic's own writeup on building effective agents. full breakdown in the article below.

6 agent patterns for AI engineers: (explained with usage) 1) prompt chaining → split the task into fixed steps, each one checking the last. → use when the task decomposes cleanly and accuracy matters more than latency. 2) routing → classify the input first, then send it to the model or tool built for it. → use when inputs fall into distinct classes that need different handling. 3) parallelization → run several calls at once and merge them, either by splitting the work or voting on the same question. → use when subtasks are independent, or when one answer deserves several opinions. 4) orchestrator-workers → a lead model decides what the subtasks are at runtime, then delegates them. → use when you cannot list the steps in advance. 5) evaluator-optimizer → one model writes, another grades, the loop repeats until it passes. → use when you have clear criteria and iteration measurably helps. 6) autonomous agent → no fixed path. it plans, acts, reads feedback from the environment, and decides when it is done. → use when the steps are unknowable and you can afford the cost and the blast radius. the first five are workflows: you wrote the path. only the last one writes its own. most production systems people call agents are pattern 1, 2, or 5 with good error handling. this taxonomy is from Anthropic's own writeup on building effective agents. full breakdown in the article below.

Hanako

143,182 次观看 • 12 天前

From -$500 in debt to $106K profit. Three weeks. 95% win rate. I wrote a Python script to scan Polymarket wallets. This one broke every filter and landed at the top of anomalies. First time I pulled up the trade history I thought the data was corrupted. Hundreds of orders in one candle. That does not happen unless something is broken. Checked the timestamps. Ran it twice. Not broken. Just insane execution. Here is what the algorithm actually does. → Account: First five minutes of every window the market is chaos. Nobody knows which way BTC is heading. Spreads blow wide open. The bot thrives here. Buys YES at 47 cents. Buys NO at 48 cents. Sells both back seconds later. Tiny profit each time. It is not betting. It is market making. Collecting pennies from everyone else's confusion. Then minute five hits. BTC trend shows up on Binance. Spot price moves. The algorithm sees certainty and changes modes instantly. Dumps the losing side. Loads the winner. Hard. 120 shares. 500 shares. Chases it all the way to 93 cents. By close it is sitting on 7,000 shares of the right outcome. Two strategies in one script. Chaos harvester for the first half. Trend rider for the second. Switches the moment the math flips. Started in the red. Ended at $106K. The equity curve does not dip. It just climbs. Stop watching candles. Start watching who is eating the spread before the move even happens.

From -$500 in debt to $106K profit. Three weeks. 95% win rate. I wrote a Python script to scan Polymarket wallets. This one broke every filter and landed at the top of anomalies. First time I pulled up the trade history I thought the data was corrupted. Hundreds of orders in one candle. That does not happen unless something is broken. Checked the timestamps. Ran it twice. Not broken. Just insane execution. Here is what the algorithm actually does. → Account: First five minutes of every window the market is chaos. Nobody knows which way BTC is heading. Spreads blow wide open. The bot thrives here. Buys YES at 47 cents. Buys NO at 48 cents. Sells both back seconds later. Tiny profit each time. It is not betting. It is market making. Collecting pennies from everyone else's confusion. Then minute five hits. BTC trend shows up on Binance. Spot price moves. The algorithm sees certainty and changes modes instantly. Dumps the losing side. Loads the winner. Hard. 120 shares. 500 shares. Chases it all the way to 93 cents. By close it is sitting on 7,000 shares of the right outcome. Two strategies in one script. Chaos harvester for the first half. Trend rider for the second. Switches the moment the math flips. Started in the red. Ended at $106K. The equity curve does not dip. It just climbs. Stop watching candles. Start watching who is eating the spread before the move even happens.

Marlow

37,951 次观看 • 6 个月前

💥Simple guide on ink Training Ground ⛓️It's L2 from Kraken ❌Kraken CEX is one of the top CEX without a token 👇Week2 Steps 🔗 ✅Join guild 1️⃣Mint 'Ink the future' article 🔗 ⛽$2.5 2️⃣Have 3 txns on InkChain ✅If followed my previous guide u can do txns ✅If trying ink for first time use bridge l 🔗 3️⃣Follow Ink on Farcaster 📝Now fill form 🔗 ✅Go to Discord ✅Go to #the-scribes-legacy-submission group ✅U will find many posting riddle ✅Copy anyone's answer and submit in discord 📝Go back to form & fill same answer in first answer 👇Watch video guide below, if u r confused 📝Quiz - Ethereum, EVM, Optimism Superchain, December 18, 2024 ✅Submit form 💥Done 💙Like 🔁RT

💥Simple guide on ink Training Ground ⛓️It's L2 from Kraken ❌Kraken CEX is one of the top CEX without a token 👇Week2 Steps 🔗 ✅Join guild 1️⃣Mint 'Ink the future' article 🔗 ⛽$2.5 2️⃣Have 3 txns on InkChain ✅If followed my previous guide u can do txns ✅If trying ink for first time use bridge l 🔗 3️⃣Follow Ink on Farcaster 📝Now fill form 🔗 ✅Go to Discord ✅Go to #the-scribes-legacy-submission group ✅U will find many posting riddle ✅Copy anyone's answer and submit in discord 📝Go back to form & fill same answer in first answer 👇Watch video guide below, if u r confused 📝Quiz - Ethereum, EVM, Optimism Superchain, December 18, 2024 ✅Submit form 💥Done 💙Like 🔁RT

CryptoTelugu

164,965 次观看 • 1 年前