← Back to leaderboard
riyazmd774

Md Riyazuddin

Verifiedactive
@riyazmd774

185k+ LinkedIn • Digital entrepreneur sharing insights on AI, tech tools online business, and Marketing. DM for Collaboration ✨ ✉️ riyaz774ai@gmail.com

RosterAI Agents & Productivity ToolsAI Content Creation & Video Gen
111,055 followers · 2,572 following · Last polled 3h ago
Perf. Score
67
Rank #89
Value Score
85.7
#38 of 273 priced
Median views
52K
p25 24K · trailing 7d
Followers
111K
-0.1% 7d
ER (impr.)
0.29%
eng ÷ impressions
ER (foll.)
0.14%
eng ÷ followers
Est. CPM
$1.16
QT-rate basis
WoW views
-8.2%
flat

Rates & value · est. from 52K median organic views

underpriced
Quote tweetbasis
$60
$1.16 / 1K views
Post
$110
$2.13 / 1K views
Thread
$155
$3 / 1K views
Retweet
$25
no CPM
Value Score 85.7 · rank #38 of 273 priced$326.09 / 1K engagementsrate age unknown (pre-audit import)

Trends

Follower growth
Median views / day
Engagement rate
View distribution · trailing-7d posts

Each blue dot is one organic post. The Performance Score ranks by the median (52K), not the mean (55K) — so a single viral post doesn't inflate a creator's reach. The p25 floor (24K) is the level 75% of posts clear.

Recent posts · ranked by views

PostPostedViewsLikesRepostsRepliesQuotesBookmarksER
Up to 60% off CapCut plans starting Oct 20. Desktop only. The Jean Phil trend made one thing clear: A memorable character can become a marketing asset. So I tested: Can an AI Video Editor help build an AI influencer? Using Codex video editing: Codex calls CapCut PC. The character gets prepared. The white model gets rendered. The campaign video gets created. Then creators refine the final story inside CapCut PC. #capcutpc #capcutchatgptplugin #capcutcodex #capcutmarketing #capcutsalefrozen9d ago168K24762270140.21%
Thinking about upgrading? Save 60% on CapCut Pro or 40% on Ultra, Oct 20–30. Desktop only. A Jean Phil-inspired character could make a product campaign stand out. But what if the video looks great overall, except for one scene that doesn't quite work? That's where editing after AI generation becomes important. #capcutpc #capcutblackfridayfrozen7d ago144K
220
33
13
0
16
0.20%
When trying a new creative tool, I care about two things: what I can actually make and how much room I have to experiment. Dreamina Basic is $1.50 for the first month, with an allowance that can cover up to 9 ten-second Seedance videos or 393 GPT Image 2.5 images. Seedance 2.5 is also part of the Seedance lineup. A pretty approachable way to explore the workflow. https://t.co/hBIPvRIutw #Dreamina #DreaminaPartnerfrozen4d ago115K20341140210.24%
The AI race isn’t just about building smarter models anymore. It’s about how much intelligence you can get for every dollar. That’s what caught my attention about MiMo V2.6 Pro, which just officially launched. A few numbers stood out: → 46 on the Artificial Analysis Intelligence Index → $0.13 weighted average cost per AA Intelligence Index task → Ahead of Qwen3.8-Max, Kimi K3 and GLM-5.3 → Just 1 point behind GPT-5.6 Sol Interesting timing too: Grok 4.7 launched the same day and also scores 46 on Artificial Analysis. But AA puts its cost at roughly $2.73-$3.74 per task versus $0.13 for MiMo V2.6 Pro. Similar benchmark intelligence, very different economics. And compared with its own previous generation, MiMo V2.6 reportedly delivers nearly 2× the intelligence of V2.5 at the same price. One workflow I tested: “Interactive 3D Solar System, with animated planets, moons, rings, 360° controls, zoom/pan, clickable planets, cinematic views and mobile support.” That single task tested coding + reasoning + visuals + interaction in one workflow. And honestly, the result was incredible.frozen1mo ago113K22458390150.30%
What if you could steal the camera move without stealing the video? Dreamina's Extract Motion function makes that possible. Prompt tools hand you a clip you can't control. A canvas hands you a project you can. Dreamina 2.0 is canvas-first - the new way AI video is made. #Dreamina #DreaminaPartnerfrozen24d ago110K19645140180.25%
Turns out the hardest part of AI video was never the image. It was getting the camera to move the way I imagined. Dreamina Canvas has Extract Motion now — feed it a video, it pulls the motion and spatial depth, and gives me a better starting point. With a full month of Dreamina Basic for $1.5 — up to 90% off, till Oct 9. #Dreamina #DreaminaPartnerfrozen27d ago110K18129180200.22%
I finally realized I was paying for the wrong part of the process. 4K on Seedance 2.5 now costs less than 720p! Reroll at 480p, then upscale only the keeper to 4K. Creative Upscale, only on Dreamina. Here's exactly how 👇 #Dreamina #DreaminaPartner https://t.co/wz2BPDDc1Qfrozen29d ago105K16325190150.21%
CapCut Black Friday is LIVE! Get 60% OFF Pro and 40% OFF Ultra, exclusively on CapCut Desktop from Oct 20–30. If you've been waiting to level up your video editing game, this is your moment. #capcutpc #capcutblackfriday https://t.co/0ULqA3jwU32d ago105K18249190180.26%
I used 3 AI agents to build a winter holiday marketing campaign. One researched the best winter destinations, one created the social media strategy, and one turned everything into a website. All working together inside Teamily 2.0. Here’s how I turned one idea into a complete marketing plan with an AI team. 🧵👇 #ad @Teamily_AIfrozen12d ago103K20449120160.27%
CapCut Pro is 60% OFF, plus 40% OFF Ultra! Oct 20–30, exclusively on CapCut Desktop. Want to turn Tesla Roadster footage into a sleek automotive hype video? CapCut PC helps you take those raw clips and give them a polished, cinematic feel. Different shots can have completely different lighting and colors. Use color adjustment tools to refine the look, balance your visuals, and build a more consistent style across your edit. Pair that with clean transitions and carefully timed cuts to make the Roadster stand out. Don’t just follow the Tesla trend. Make the visuals your own with CapCut PC. #capcutpc #capcutmarketing #capcutsale #capcutblackfridayfrozen4d ago102K11951220130.20%
I think I just found a productivity cheat code 👀 Tired of switching between different AI apps just to test the same prompt? WorkBuddy brings multiple AI models into one place. You can switch between GPT-6 Astra, Gemini-3.5-Flash, and DeepSeek-V4.1-Flash for coding, reasoning, creative writing, or experimenting with prompts. 🔥 Hy4 Preview is free until Oct 10 👀 And Grok is reportedly coming soon. Less tab switching. More experimenting. Give @WorkBuddy_AI a try while the 2-week free access is live!frozen13d ago93K11534200200.20%
CapCut Black Friday is live today, offering 60% OFF Pro and 40% OFF Ultra. The offer runs October 20–30 on CapCut Desktop only. My favorite moment from the day happens halfway through the footage. It could make a great opening. In CapCut PC, I can lead with that highlight, return to the setup, and build the vlog around the moments that matter. Captions, music, and carefully timed transitions help connect the sequence. Give your everyday footage a story shape that makes the best moments easy to find and enjoy. #capcutpc #capcutblackfriday #capcutsale #capcutmarketing1d ago86K20149170110.32%
Step 5 Preview just hit #1 on OpenRouter’s LLM leaderboard. 👀 Recently, I gave it a screenshot and a real product requirement. Then I said: Build it. No step-by-step instructions. No manually written code. I wanted to see if @StepFun_ai could go beyond generating code and actually complete the task. I provided a reference design, product requirements, functional interactions, and acceptance criteria. It built the interface, implemented interactions, tested the result, and fixed issues along the way. The real test wasn't the first version. It was how the model handled things that didn't work as expected. It kept iterating until the result moved closer to the requirements. That's the difference between generating code and working toward a finished product. Step 5 Preview is available on OpenRouter, Kilo Code, Cline, Hermes Agent, and OpenCode. Try it → https://t.co/O0FoTv7RQp Open weights arrive October15.frozen13d ago83K12829130170.23%
This AI subscription costs less than my coffee. ☕️ And somehow, it gets you a full month of Dreamina Basic for just $1.50. Not a 3-day trial. Not a handful of credits. A full month. Here’s what makes the offer interesting: 🎨 Up to 393 Image 2.5 generations 🎬 Up to 31 Seedance 10s videos 🔥 Up to 90% OFF The promotion runs September 23 October 9. If you’ve been looking for an inexpensive way to experiment with AI images and videos, $1.50 for an entire month is hard to ignore. 🔗 Link in the comments 👇 #Dreamina #DreaminaPartnerfrozen29d ago80K20532121150.33%
The $1.5 Basic offer has been extended and is still available. Pretty good timing to try this workflow. I kept paying for 4K drafts I never used. You won't believe this: 480p to 4K for just $1. Reroll at 480p, then upscale only the keeper to 4K. That’s Creative Upscale, exclusive to Dreamina. Give it a try right here: https://t.co/O5rvVgZpdx #Dreamina #DreaminaPartnerfrozen8d ago71K12525112170.25%
Good video editing is about making the creative process smoother, not just making content look better. Meet Filmora 16, bringing new features to make video editing easier and more efficient. 🎯 Camera Tracking — Keep moving subjects in frame without manual keyframing. 🎨 Color Mask — Adjust specific areas of your footage with greater precision. ✍️ Smart Text Correction — Fix caption errors more easily. 🎙️ Multi-Source Recording — Record your screen, camera, and audio together, then edit each separately. These practical upgrades can help simplify editing and make content creation more efficient. Explore Filmora 16 here 👇 🔗 https://t.co/d1TL2NdeKj #Filmora16 #VideoEditing #ContentCreation2d ago66K18461160180.43%
Looking to create fun Halloween content without spending hours on it? 🎃👻 I came across Wizstar, and it’s packed with creative, high-quality AI templates to experiment with different styles, spooky looks, and unique ideas. It’s easy to use, fun to explore, and perfect for getting creative this Halloween. Definitely worth checking out! 🔥frozen5d ago61K10750160110.30%
Sound on 🔊 Your AI video looks straight out of a fantasy movie… until random background music suddenly ruins the scene. 🎵 And the worst part? You didn't even ask for it. This is one of the frustrating problems with AI-generated videos. The visuals look incredible, but unwanted music can ruin the atmosphere. Regenerating the entire video isn't always the answer. Meet Mirelo SFX 2.0 by @MireloAI, a video-to-sound model that watches your footage and generates sound effects and ambience synced to the action on screen. I'm testing it on a mystical beasts video where unexpected music appears in the original footage. Here's what to listen for: 🔊 BEFORE: Unwanted music disrupts the scene. 🔊 AFTER: Mirelo removes the unwanted music and generates sound effects and ambience that match the visuals. No need to regenerate the picture just to rework the sound. From fantasy AI clips to complete video edits, this opens up some interesting possibilities for AI creators. Mirelo SFX 2.0 is free to test. Give it a try and hear the difference for yourself. 👇 https://t.co/Nufr5RdiLRfrozen8d ago57K1155050130.32%
Most AI agents wait for your next prompt. comma keeps the work moving. Give it a goal, and it can work in the background, track updates, wait for replies, browse the web, and take the next step without constant instructions. From research to tasks that take hours or days, it keeps things moving. Need access to personal accounts, local files, or payments? comma can continue on your own computer or phone when needed. Your context and progress stay with the same agent across apps and devices, so you don't have to start from scratch. That's the real value of a 24/7 personal agent: set the goal, step away, and let it get to work. Explore comma at https://t.co/xKA93qwtRAfrozen13d ago53K10716220190.31%
One thing I’ve noticed while exploring AI agents: getting live web data consistently is harder than it sounds. JavaScript-heavy websites, unreliable extraction, and scaling across hundreds of pages can quickly turn a simple workflow into a headache. That’s why Oxylabs Web API caught my attention. It combines two essential capabilities in one API: 🔎 Search — Find relevant sources across the live web. 📄 Scrape — Extract clean, structured data from web pages. With MCP Server, Agent Skills, and JSON or Markdown output, it also fits naturally into AI agent workflows. I can see this being useful for competitor research, price tracking, and agents that need fresh web data. You can try it for free and get $5 in credits for your next project. 👉 https://t.co/z6tJKR6X6V Your AI agent needs more than intelligence. It needs fresh data. #Oxylabs #AIAgents13h ago52K16942190140.47%
Most end to end tooling wants a deployed URL. For coding agents that is a timing problem more than a convenience problem. The agent works locally. If the only real test of its work runs after a push and a deploy, the result arrives after the session that wrote the code has ended. The context is gone, and somebody else, usually you, picks up the failure the next morning. The fix is to move the check inside the session. The TestSprite CLI can point Integration Tests at an app running on a local port, with no deploy and no public URL, so the agent sees the failing step and the screen at that moment while it still remembers why it wrote the code. Within 60 seconds of posting, reply to your own post with: The CLI is @Test_Sprite/testsprite-cli. Setup takes a few minutes.frozen9d ago40K9244150110.40%
Notice how every AI voice sounds the same? Gradium Studio lets you clone or design a voice for free, then run it on device with Phonon. This French-speaking dino uses one of the built-in voices 🤯 https://t.co/8RcOI1xFCm https://t.co/arspuPu6sZ2d ago34K66320120.24%
Bro… what if the real AI bottleneck isn’t intelligence? It’s that the model can’t keep what it learns. Boltzbit just published research on Bayesian Self-learning Transformers (BAST), where incoming data becomes targeted updates to the model’s own weights without retraining the entire system. Not another note in a memory store. Not another chunk added to the prompt. The model itself adapts. Now I’m curious what happens when millions of agent tasks stop becoming disposable data. 👀 Here’s the paper if you want to dig in: https://t.co/HRTQzKuAcNfrozen29d ago33K3300040.11%
The crazy part isn’t AI writing code faster. It’s that the whole team can now help build. Support gets a customer request → Base Code turns it into a PR → engineers review and approve. No local setup. No waiting for someone to translate every idea into a ticket. The codebase is finally accessible beyond the people who have it on their laptop.frozen24d ago32K44031100.18%
AI video is getting faster, but reliability is what actually saves time. I like the idea of OnSolo’s Solo Video generating the full 15 seconds in one pass instead of making you burn credits retrying the same shot. And unlimited access for campaign members? That makes it worth testing.frozen29d ago30K49020100.20%
The real test for an AI app builder isn’t how good the prototype looks. It’s whether you can take the code, own it, and actually ship it. Frontend + backend + database. Yours. That’s what makes Myndlab interesting.frozen15d ago29K54000120.22%
If you're starting from zero: it's under Create, second tab. Generate one shot, then branch that node rather than rerolling. Timeline is right there too. $1.5 gets you the full month, ends Oct 9. https://t.co/zhoGxM3lEdfrozen29d ago29K3110090.14%
The part of AI I'm most curious about isn't just getting answers. It's figuring out what I can create and automate with it. That's what made this session stand out to me. There's a live GPT 6 Astra Crash Course where you build with Astra during the session, not just watch. What you'll get: → Product building: describe the app or tool you want, and watch Astra build a working version live while you follow along → Job hunting: Astra finds jobs that match you and tailors your resume for each one → Investing: tell Astra the stocks, mutual funds and IPOs you follow, and it alerts you when they move and explains why → The Astra Power Playbook, with workflows ready to plug and play, for everyone who attends And the best part: it's completely free. Happening this Saturday, 10 AM EST. https://t.co/l6BuhvS9U2 #AIWorkflowsfrozen9d ago29K884940120.54%
Bookmark this for the next PDF full of awkward tables. 🔖 OXR: 13 types of document elements into Markdown or JSON Coordinates kept, so you can check the source region First on olmOCR-Bench, OmniDocBench and DocParsingBench Best time of any open-source parser: 101s vs MinerU2.5-Pro's 254.7s (same GPU, same 100 pages) Layout model can be replaced if your documents need different boundaries Free online demo https://t.co/khvP2hgFjU2d ago27K5130040.21%
The number I’d track with StepAudio 3 Music: time until I have something I’d actually use. That includes briefing, listening, rejecting and revising. Generation speed alone doesn’t tell you how much work a tool saves. Worth evaluating across a complete task.frozen23d ago26K45040110.23%
Agents learned to use computers, now they're getting bodies. Dyna-2.1 is an agent built around Dyna-2 on new semi-humanoid hardware.frozen22d ago24K43020120.24%
Viggle now has a model roadmap: JST-2 handles 3D characters, motion and scenes, and JST-3 replaces rule-based physics with learned physics. that's the kind of roadmap you'd expect from a model lab https://t.co/ztkOMpSGTXfrozen3d ago24K49130160.29%
This is Insane...MoE is starting to look less like just a scaling trick for datacenter models and more like a genuinely interesting architecture for constrained hardware. Sparse compute was always attractive on paper. Making the memory movement work in practice is the hard part.frozen10d ago22K39000110.22%
The crazy part isn't better voice models. It's the check after. @OnepinAI reviews each take for naturalness and pronunciation and fixes the wrong words directly, so you're not regenerating the line or hiring a voice actor.frozen21d ago22K50110110.28%
The most revealing result from World Models' Last Exam in Physics: Even the strongest models collapse on the Hard group. Collisions, motion reversals, changes of state this is where physics exposes the gap. For WAMs, wrong physics means wrong plansfrozen14d ago22K45100110.26%
Most AI app builders stop once the UI looks good. Myndlab is going further frontend, backend, database, and you own the code. Curious to see what this unlocks. October 6 👀frozen21d ago20K42000110.27%
The timing on this article is honestly too good 😂 OpenAI announces the idea, and Teamily basically says: “Yeah… we already built that.” Love when the product ships before the announcement does.frozen23d ago18K31000100.23%
Moving our VR prototype up the schedule for the Chroma Awards. Core loop was built in April so it's eligible, it just needs a second level3d ago18K3400030.21%
Pairing my AI video clips with Eleven v4 dialogue this week. The free access couldn’t have come at a better time for my festival submission.frozen21d ago18K32000110.24%
Coding agents are getting good at writing code. The harder test is whether they can actually see what they’re building. PPTBench puts that to the test: understand a visual target → reconstruct it in editable code → render it → inspect it → fix what’s wrong. 18,000 reconstructions later, only 2.57% passed every gate with zero recorded defects. The gap between “looks close” and “is actually correct” is still huge. That’s where visual coding gets really interesting.frozen23d ago17K35050120.31%
My product render spin got a whoosh and a soft hum.frozen6d ago15K52000110.41%
Been building with Ultravox for a while, and now my agents just sound way better without me changing a thing. Pretty cool upgrade.frozen22d ago15K33010120.30%
an impossible interview where every round is tougher than the last. whoever's at the top of this, what were they asking you by the end?frozen6d ago15K34010100.30%
A household task won’t stay perfectly tidy while it’s happening. Replanning feels like part of the job, not some weird edge case. https://t.co/lYvXpiM2e3frozen10d ago13K3200140.29%
Original expression is a hard promise to keep. Seeing a child’s silly idea honored proves this isn't just generating boilerplate text.frozen9d ago12K3310030.30%
$5K. 3 weeks. One product video. 😳 Or… 5 minutes. One screen recording. Done. PuppyDog turns your actual product into a polished video with the script, voiceover, animations, and editing handled by AI. No generic avatars. No weeks of back-and-forth. No expensive agency. This is what AI-powered product marketing should feel like. 🔥frozen29d ago12K32010110.36%
Sound on 🔊 Your AI video looks straight out of a fantasy movie… until random background music suddenly ruins the scene. 🎵 And the worst part? You didn't even ask for it. This is one of the frustrating problems with AI-generated videos. The visuals look incredible, but unwanted music can ruin the atmosphere. Regenerating the entire video isn't always the answer. Meet Mirelo SFX 2.0 by @MireloAI, a video-to-sound model that watches your footage and generates sound effects and ambience synced to the action on screen. I'm testing it on a mystical beasts video where unexpected music appears in the original footage. Here's what to listen for: 🔊 BEFORE: Unwanted music disrupts the scene. 🔊 AFTER: Mirelo removes the unwanted music and generates sound effects and ambience that match the visuals. No need to regenerate the picture just to rework the sound. From fantasy AI clips to complete video edits, this opens up some interesting possibilities for AI creators. Mirelo SFX 2.0 is free to test. Give it a try and hear the difference for yourself. 👇 https://t.co/vbaGhjRcBB8d ago8.8K181616010.58%
The best part of inventing a product is deciding it has a wildly specific problem to solve.frozen13d ago8.0K23000110.43%
https://t.co/TaWdwaQkPP fans: Bob or Labubu? I'm leaning Bob, mostly because of the goggles. You can actually do something with your pick in Meme Arena: supporting a fighter helps charge their ultimate. Vivix W1 powers this endless interactive AI livestream with cheaper, faster generation. Go back your side. I suspect the ears are going to win a lot of people over.frozen17d ago7.3K2900050.46%