AJy 🎨👨‍💻

How i shipped a custom architecture for a deep research ai agent

@prayag_sonar @Google @samissudo

@Google Gemini btw

Today we’re expanding the Gemini family with three new models built to be faster, more token efficient, and reliable at scale. Meet the new Gemini models ↓

@CarlosBBuild @athcanft i know someone who will

@athcanft I have a skill issue, if I gave you $500, would you make an ad for me?

@0xROAS I was just so younga like 🫳

@tetsuoai @samissudo wonder if you can use modal with this

Clone any voice. Keep it local. Voice Clone Lab turns 5–15 minutes of audio into a voice model on your own GPU, cleanup, transcription, fine-tuning, and generation included. Open source: github/tetsuo-ai/voice_clone_lab

@samissudo @Richelle_Ji Time to get rich mehn , can one use this rn ?

@Richelle_Ji WTF @johnama0

Played with my Lucy 2.5 date's hair, live. and our joint arms. 🤯you can live stream an ai-generated subject/object. more about how it works ⬇️

@tivollem @samissudo can’t we consider this demand ?

Je peux générer 100 créas vidéo en moins d'1h. Avec Claude Code. Sans CapCut. Sans monter une seule vidéo à la main. Il me sort directement les créas prêtes à tester. Hooks. Sous-titres. Musiques. Variantes. Exports. J'ai documenté tout mon workflow. Commente "CLAUDE" je te l'envoie. (oui la vidéo est en vitesse réelle)

@Vicryptweb How much was your ad spend for this campaign ?

I generated 3,000 leads and did N1,000,000 in the last few days. Using the Telegram DM funnel strictly Here is how I did it (This knowledge is worth N50,000 but I'll share it here for FREE) Make sure you bookmark this post so you can always come back for reference. Also repost this so other Digital marketers can see. The Telegram DM funnel still works if you can solve these 3 problems 👇 1. Reducing the steps friction to minimize drop offs- This means people not completing the steps before seeing your offer because of many things to do. 2. A medium for following up on leads later 3. Getting them to save your number- This is important as well because if they save your number their story section will feature you and they also get notified when you post a story Otherwise they can only see your story when they see your chats or profile. 4. Creating an urgency/Fomo for them to buy immediately not when they need it. When I decided to test this funnel, I saw these problems, finding solutions was what gave me results. So let's break it down 1. To solve the issue of drop offs, people not completing the steps before seeing the offer You will need to sound more human with your Auto response. I don't advice using Emojis. Use the multiple message feature. There is also a feature To mention their first name or last name to make it more personal. In your Tel Auto responder it is called Tag To break your texts. Add a space of 3-5 secs between texts. Nigerians don't like bots to respond to them, so you need to sound more human. 2. You need them to Join a community first. Don't tell them to save your number and type saved NO 👎 This is the normal step process people use Ad→DM→Save number→Join community →Get Vsl/sales page The issue with this is many don't complete the process because it is long. Use this instead. Ad→DM→Join community →Get Vsl/sales page→Save number. This works better because telling them to save your number first before anything either makes them just prompt for next response like saying "Saved" without actually saving. Because they don't have all that time to waste And you've not also proven your offer to be valuable to them. Giving them a community to join for more information without any efforts first is better. As they Join they prompt the next response from your Auto responder. Which gives them Your offer. A VSl or a sales page. Use a sales page with a 5 minutes video summary of your offer along side a detailed written explanation on the page. If you must use a VSL. Make it short, 15 minutes is okay, don't make it long 30 minutes looks like a chore. Don't say serious people will watch it. Its not about that its about the average phone user having a low attention span. 3. To get them to save your number this is what I do. After I've given them the offer I set multiples messages to 15 minutes (900 secs) so 15 minutes after they have seen my sales page or VSl Then I tell them why they should save my number. Now they have seen a reason to save it. I set other multiple messages to do some quick follow up like Hope you've seen this and you understand, is this a problem you have and do you think this is the solution? 4. This is how I create an urgency/fomo Using this same multiple message at (900 secs) after it has sent the quick follow up message After another 15 minutes (900 secs) its send a message telling them my actual price and the discounted price Also telling them why they should key in because of limited slots. After another 15 minutes another follow up message drops Now this follow up message has a link to a particular story posted about my offer. This triggers them to see my story if they haven't. My story lasts for 45 hours so every 45 hours I update a new story and change the link in the auto responder. This is the funnel I used.. Follow me @Vicryptweb for more updates like this Check my pinned tweet to learn how i do 2k-5k leads consistently

@samissudo @crimson_moti Yeah this lip sync is good though

@johnama0 @crimson_moti yea it can, it's just that lipsync

THIS GIRL IS LYING ON A WHITE PILLOW WITH MESSY HAIR AND FRECKLES AND SOMEONE CHARGES $800 TO BUILD HER. She wakes up. Messy blonde hair on a white pillow. No makeup. Wrinkled white t-shirt. She picks up the phone, looks into the camera half asleep, and asks "can you tell?" Then the camera pushes in. Close-up. Freckles. Pores. The redness around her nose. The kind of skin detail you only see in a mirror first thing in the morning. Every single pixel of it is AI. Pause at 0:05. Look at the skin. That's not a filter. That's not a texture pack someone dragged onto a face. Those are individual pores and freckles that the system generated because the negative prompt was set up before the render. Most AI faces can't survive a close-up like that. This one leans into it. I'm not saying AI faces are new. Calm down. We've all seen the perfect symmetrical doll faces that fool nobody. But this is not that. This is a "just woke up" selfie with bed-head and skin imperfections and the look on her face when she hasn't had coffee yet. That's the part that should stop you. Nobody told the AI to add freckles. The system did it because the skin texture pass was set up correctly before the face was ever animated. HERE'S WHAT MOST PEOPLE MISS ABOUT WHY THIS CLIP WORKS. The creators who generate perfect AI faces wonder why nobody engages. It's because perfect is the tell. The human eye doesn't trust symmetry. It doesn't trust flawless skin on a morning selfie. It looks for the things that shouldn't be there if the face was generated: the freckle, the pore, the messy strand of hair covering one eye, the slight redness on the nose. This clip has all of them. That's not luck. That's a negative prompt node running 40+ exclusion terms that strip the "AI look" before the face is ever rendered. I build AI personas and the mechanic underneath this clip is the one thing I tell every creator who asks why their content looks fake. You didn't break the face. You skipped the skin texture pass. You sent a clean symmetrical render straight into animation and the audience clocked it as AI in the first second. This girl has freckles on a pillow because someone ran the negative prompt correctly and refused to animate until the skin passed the close-up test. Pause at 0:08. She touches her chest and says "this is AI." By that point most people have already forgotten they're watching a generated face. That's the whole game. The freckles did the work in the first five seconds and after that the brain stops questioning. The creators adding imperfections on purpose are the ones whose personas survive past week one and clear $800 per character. Not because imperfection is a style choice. Because imperfection is what makes a human brain stop analyzing and start believing. And the moment someone believes the face they follow it and they pay for it. Watch it twice. Count the freckles at 0:05. That's the pipeline nobody talks about.

@samissudo @twoclipping 👀

@twoclipping right

gpt 5.6 + seedance 2.0 is a f*cking cheat code swapped fable out of my ugc system for it and the whole thing runs 10x cheaper and faster now proof: one of these girls is a real creator. the other is the ai rebuild. same script, same visual hooks you won't clock the fake inside the system im running: - the character ref prompts (named imperfections stack) - the video prompts + the anti ai negatives - the one line fix every gpt 5.6 script needs - the full rebuild workflow reply LEFT or RIGHT and i'll dm you the answer + the full prompting system (follow so i have dm access)

@nickscamara_ You guys have a features that monitors the web ? What do you mean please I could need it

Monitor the web with Firecrawl + Grok 4.5

@davidfigeira Want

@twoclipping Prompts

fable 5 is INSANE for ugc ads i put every prompt behind our hyper realistic ai ads into one skills library. built from 1,855 viral ads scanned, 1,204 rejected, 61 winners reverse engineered every girl in this video came out of it. not one of them is real. under $1.20 an ad, under 15 minutes inside the library: the character lock that holds one face across every cut the voice formula that never drifts between scenes the anti ai negative stack that kills the "ai look" the full 8 beat ad template. fill the brackets, render a ugc creator charges $300-500 for ONE of these videos comment "PROMPTS" and ill dm you the full library (follow so i can dm)

@RaenestApp Hey, i sent a message in the app and for more than 24hours now no response. i have shared the issue in DM, please check

A couple of weeks ago, we rolled out a major update to how you receive USD payments on Raenest, with a new banking partner built for faster payouts and long‑term stability. You shared a lot of questions and feedback. This thread provides clear answers to those questions.

@folaoftech @samops

The engineering trick behind GTA's seamless open world.

@Vicryptweb review

There are two problems every Digital marketer must solve 1. Funneling 2. Lead generation You can have a good funnel with a bad lead game. You can have a good leads game with a bad funnel I've experienced both this year. Always review your funnel and ask other people to help you review. Comment review and send me an Inbox I'll help you review yours now

@RaenestApp promptly ? you people have still not responded

@johnama0 Hi There, We apologise for the delayed response to your messages. Please share your registered email address via DM so we can assist promptly.

@RaenestApp I've sent a message from 2 days now, no response, can you guys just respond ? ! i have also shared my email in dm , look into it please!

$1,000 in SanDisk in January would be worth $7,937 right now. Here’s what a $1,000 investment in January would look like today across 10 of the year’s strongest stock performers. Which of these is in your portfolio? 👀 Not financial advice.

50% profit on $1,000 is $500. 5% profit on $100,000 is $5,000. Chase bigger capital not bigger returns.

@RaenestApp @RaenestApp please respond , I have some funds that was sent to that lead bank account and it has not landed since

@RaenestApp Hey, i sent a message and for more than 24hours now no response.