Hadi (@metaways_) header
Hadi (@metaways_) profile picture

I make AI-powered videos + deep-dive YouTube content. Exploring tech, creativity & the future.

Joined January 2014
Photo by Hadi (@metaways_)Photo by Hadi (@metaways_)Photo by Hadi (@metaways_)Photo by Hadi (@metaways_)Photo by Hadi (@metaways_)Photo by Hadi (@metaways_)Photo by Hadi (@metaways_)Photo by Hadi (@metaways_)Photo by Hadi (@metaways_)
πŸ“‚ GTM ┃ ┣ πŸ“‚ RevOps ┃ ┣ πŸ“‚ 1 CRM as the record ┃ ┣ πŸ“‚ Signal routing to owners ┃ ┣ πŸ“‚ Scheduled enrichment ┃ ┣ πŸ“‚ Weekly deal reviews ┃ β”— πŸ“‚ Loss reasons ┃ ┣ πŸ“‚ ICP and TAM ┃ ┣ πŸ“‚ ICP from closed-won deals ┃ ┣ πŸ“‚ Segments ┃ ┣ πŸ“‚ Personas ┃ ┣ πŸ“‚ Lookalikes ┃ β”— πŸ“‚ 1 account list ┃ ┣ πŸ“‚ Signals ┃ ┣ πŸ“‚ Hiring ┃ ┣ πŸ“‚ Funding ┃ ┣ πŸ“‚ New hires and GTM leaders ┃ ┣ πŸ“‚ Past champions ┃ ┣ πŸ“‚ Tech stack ┃ ┣ πŸ“‚ Public records ┃ ┣ πŸ“‚ Website visitors ┃ ┣ πŸ“‚ Review sites ┃ ┣ πŸ“‚ Post engagers ┃ ┣ πŸ“‚ Target-account exec posts ┃ β”— πŸ“‚ Brand mentions ┃ ┣ πŸ“‚ Scoring ┃ ┣ πŸ“‚ Fit score ┃ ┣ πŸ“‚ Intent score ┃ ┣ πŸ“‚ Tiers 1, 2, 3 ┃ ┣ πŸ“‚ Score decay ┃ β”— πŸ“‚ Hot, warm, cold routing ┃ ┣ πŸ“‚ Data and enrichment ┃ ┣ πŸ“‚ Data sources ┃ ┣ πŸ“‚ Email waterfall ┃ ┣ πŸ“‚ Email verification ┃ β”— πŸ“‚ Mobile numbers for the top tier ┃ ┣ πŸ“‚ Outbound ┃ ┣ πŸ“‚ Sending infrastructure ┃ ┃ ┣ πŸ“‚ Secondary domains ┃ ┃ ┣ πŸ“‚ Warmed inboxes ┃ ┃ β”— πŸ“‚ SPF, DKIM, DMARC ┃ ┃ ┃ ┣ πŸ“‚ Copy ┃ ┃ ┣ πŸ“‚ First line per signal ┃ ┃ β”— πŸ“‚ 1 test per round ┃ ┃ ┃ ┣ πŸ“‚ Channels ┃ ┃ ┣ πŸ“‚ Email ┃ ┃ ┣ πŸ“‚ LinkedIn ┃ ┃ β”— πŸ“‚ Calls for the top tier ┃ ┃ ┃ β”— πŸ“‚ Reply routing ┃ ┣ πŸ“‚ Reply labels ┃ β”— πŸ“‚ Approval gate ┃ ┣ πŸ“‚ Inbound and pipeline ┃ ┣ πŸ“‚ Inbound lead enrichment ┃ ┣ πŸ“‚ Meeting prep and follow-up ┃ ┣ πŸ“‚ Product sign-up outreach ┃ ┣ πŸ“‚ Webinar and event attendees ┃ β”— πŸ“‚ Closed-lost revival ┃ ┣ πŸ“‚ Content ┃ ┣ πŸ“‚ Founder and team posts ┃ ┣ πŸ“‚ Buyer questions from calls ┃ ┣ πŸ“‚ Weekly newsletter ┃ ┣ πŸ“‚ Live sessions and clips ┃ β”— πŸ“‚ Gated templates ┃ ┣ πŸ“‚ ABM and paid ads ┃ ┣ πŸ“‚ Target-list audiences ┃ ┣ πŸ“‚ Promoted team posts ┃ ┣ πŸ“‚ 1-to-1 ads ┃ ┣ πŸ“‚ Intent in the account score ┃ ┣ πŸ“‚ Gifting ┃ ┣ πŸ“‚ Retargeting ┃ β”— πŸ“‚ Search ads ┃ ┣ πŸ“‚ Partnerships ┃ ┣ πŸ“‚ Warm intros ┃ ┣ πŸ“‚ Customer referrals ┃ ┣ πŸ“‚ Revenue share ┃ ┣ πŸ“‚ Integration listings ┃ β”— πŸ“‚ Shared accounts ┃ ┣ πŸ“‚ Community ┃ ┣ πŸ“‚ Podcasts ┃ ┣ πŸ“‚ Dinners ┃ ┣ πŸ“‚ Launches ┃ ┣ πŸ“‚ Paid community ┃ β”— πŸ“‚ Slack and Discord groups ┃ ┣ πŸ“‚ AEO ┃ ┣ πŸ“‚ Prompt tracking ┃ ┣ πŸ“‚ Answer-first pages ┃ ┣ πŸ“‚ SEO basics ┃ ┣ πŸ“‚ Customer reviews ┃ β”— πŸ“‚ Forum answers ┃ ┣ πŸ“‚ AI agents ┃ ┣ πŸ“‚ Lead scoring ┃ ┣ πŸ“‚ Pre-call research ┃ ┣ πŸ“‚ Signal to sequence ┃ ┣ πŸ“‚ Copy intelligence ┃ β”— πŸ“‚ Meeting action items ┃ β”— πŸ“‚ Reporting ┣ πŸ“‚ Reply rate ┣ πŸ“‚ Positive reply rate ┣ πŸ“‚ Show rate ┣ πŸ“‚ Spam and bounce rate ┣ πŸ“‚ Meetings booked per signal β”— πŸ“‚ Pipeline by motion
14
19
6
173
17,167
cancel your clay subscription here is an open source version it has - waterfall enrichment - postgres databases for hosting - your coding agent can run it - you can build a TAM map on it - you can build a signal tracker with it for your TAM - it's multiplayer with your whole GTM team, and you can make it read only for the normies and you can get it right now get started for free - graphed.com/ and here's the skill file - github.com/graphed-inc/marke…
"But like.. what does @clay even do!?" -- Here's my take. Explain Clay to: 1. My Grandma: Sales AI platform 2. A first Date: AI assistant for sales & marketing. Research potential customers, draft emails, handle repetitive work so people have more time for real conversations (and life). 3. My art school friends: Businesses buy contact data to figure out who to prioritize. Clay is like Netflix for those customer insights + Agents that help make sense of it to automate the busywork 4. My ex-finance friends: Excel with 220+ plugins like FactSet or CapIQ with concierge account Agents. Enrich prospects with custom firmographic meta-data. Agents route prospects based on context and tier + execute build, launch, and iterate campaigns. Lower CAC. Raise deal velocity. 5. To CROs: Clay is a GTM orchestration and enrichment platform, combining your 1P data with 220+ data providers to give you the maximum coverage and accuracy. Unify your sales & marketing team with this always-updating source of truth. Build inbound plays, outbound automation, rep productivity, and ABM campaigns to turn your revenue systems into a self-learning GTM motion. β‡’ See how you can replicate the same systems build by Anthropic, OpenAI, SpaceXAI: dub.sh/uDlnkKw 6. To GTMEs & RevOps: Data hygiene is solved. Now, orchestrate your GTM data and build workflows straight from your agent of choice. Get the benefit of enterprise-grade infra (scale, audibility, security, credit budgets, etc) along with the "idea-to-execution" agility of your best agentic plays. What'd I miss? (Shout out to grandma CROs everywhere).
33
17
355
29,105
"B2B SaaS, 50-500 employees, VP of Sales" is 3 answers. A full ICP has 34. We went through 1M+ cold emails in @InstantlyAI. Tight lists of 500-1,000 people with a reason to reply got 20-30% reply rates, 100K blasts got 2-3%... and your ICP is how you know who has that reason. So I broke it into 4 parts (every question is in the infographic, feel free to save it): 1. Company fit Our example: you sell a tool that helps SDR teams book more meetings. Before opening any tool, export 2 lists from your CRM: the companies that paid you, and the ones that paid you then left. Every answer gets checked against those 2. We use folk as our CRM, plugged into Claude Code, so pulling both is 1 prompt. Data tools answer the first 6 questions: β†’ Limadata for what they sell, to whom & where β†’ Prospeo for industry, headcount & location β†’ ColdIQ company search for funding stage, with the hiring filter on to catch the teams that are growing β†’ BuiltWith for the tools running on their site The last 3 (the job they need you for, what's blocking their growth, what your churned accounts share) come from your deal notes. Then run your Prospeo & ColdIQ filters against your won list. If a best customer drops out, one filter is too tight. Whatever your churned accounts have in common becomes a "do not target" line, which in our example is teams with only 1 SDR. 2. User persona The user feels the problem first, so most of your messaging is written for them, starting with the number their boss checks. An SDR Manager running 4 reps is judged on meetings booked per rep, and also cares about pipeline generated & how fast new hires ramp. Every email to them should hit one of those 3. A Head of Growth running 2 people can be a second user. A Marketing Intern working alone is out, since they don't run a team or push a purchase. Then pull the exact phrases they use to complain from your call transcripts, and reuse them word for word in your emails & posts. 3. Buyer persona The buyer is rarely the user, so map every seat that has to say yes: the SDR Manager brings you in, the VP of Sales signs, Finance checks the budget... and IT or Legal can still kill it. Apollo or Prospeo find who sits in each seat (domain + job title in, names out). Do it on 3 of your won deals and write down the proof that got each one signed: a case study, a trial, a reference call. 4. Signals Visitor ID in ColdIQ (a pixel on your site) shows which companies hit your pricing page. Job-change data from Signalbase & LeadMagic flags past users who moved to a new company. PredictLeads picks up hiring, funding & new tools, with the date each tool was first and last seen. But the signal worth finding first is in your won deals. Go back through them, check what happened at each company right before the deal started, and watch whichever signal keeps showing up. That said, a signal on its own only means the timing is good. An account only becomes Tier 1 when the company passes Part 1, your user works there & you can name who sits in each buyer seat. Those get your first emails: open on the signal that fired, tie it to your user's number & add the proof your signer needs. And if a company fails Part 1, it stays out (even with a strong signal). P.S: the full guide is below, with the Claude Code prompt for each part + a 7-day plan that ends with your first Tier 1 list :)
25
5
49
2,594
a masterclass on writing cold email offers that book sales calls without a guarantee or a single generic testimonial, in ANY b2b niche: (bookmark)
2
1
19
492
Editing inside Claude is getting really good. Really fast. But don't just expect to hand them some footage and have them create a banging ad for you. Here's what we've been doing: 1/ Give it a detailed brief. We've been handing it the same brief our normal editors work from. So Claude has all the information it needs for that video. 2/ Hook your Claude up to some kind of shared brain like Runneth, HQ or a shared GitHub Repo. That way, all the learnings compound, and skills can be shared, etc. 3/ Create skills from your past winners. Give the finals to Claude, ask it to review the videos and understand the editing style. Then these skills can be added forever. 4/ Make your team use it. Yes, force them. Otherwise, editors just won't use it. This is a way for them to speed up and remove the boring work like first drafts. 5/ Use XML files. Sometimes it is just faster to make the small tweaks yourself. So set up a skill that delivers the finished video along with an XML that editors can import into Prem to make polish edits, etc. 6/ Keep training it. Don't stop at it's first try, just keep building on the skills to make it 10x better.
14
5
82
6,499
Hadi retweeted
πŸ“‚ content marketing ┃ ┣ πŸ“‚ research ┃ ┣ πŸ“‚ outlier videos ┃ ┣ πŸ“‚ competitor accounts ┃ ┣ πŸ“‚ content trends ┃ ┣ πŸ“‚ trending sounds ┃ β”— πŸ“‚ reading comments ┃ ┣ πŸ“‚ hooks ┃ ┣ πŸ“‚ spoken hooks ┃ ┣ πŸ“‚ on-screen text hooks ┃ ┣ πŸ“‚ visual hooks ┃ ┣ πŸ“‚ hook bank ┃ β”— πŸ“‚ first 3 seconds ┃ ┣ πŸ“‚ formats ┃ ┣ πŸ“‚ talking head ┃ ┣ πŸ“‚ green screen ┃ ┣ πŸ“‚ slideshows ┃ ┣ πŸ“‚ POV skits ┃ ┣ πŸ“‚ day in the life ┃ β”— πŸ“‚ product demos ┃ ┣ πŸ“‚ angles ┃ ┣ πŸ“‚ pain points ┃ ┣ πŸ“‚ before & after ┃ ┣ πŸ“‚ myth busting ┃ β”— πŸ“‚ founder story ┃ ┣ πŸ“‚ scripts & briefs ┃ ┣ πŸ“‚ creator briefs ┃ ┣ πŸ“‚ shot lists ┃ ┣ πŸ“‚ script templates ┃ β”— πŸ“‚ brand voice ┃ ┣ πŸ“‚ AI content ┃ ┣ πŸ“‚ Claude Projects ┃ ┣ πŸ“‚ content agents ┃ ┣ πŸ“‚ AI UGC ┃ β”— πŸ“‚ AI slideshows ┃ ┣ πŸ“‚ distribution ┃ ┣ πŸ“‚ TikTok ┃ ┣ πŸ“‚ Instagram Reels ┃ ┣ πŸ“‚ YouTube Shorts ┃ ┣ πŸ“‚ posting schedule ┃ β”— πŸ“‚ repurposing ┃ ┣ πŸ“‚ UGC & creators ┃ ┣ πŸ“‚ creator sourcing ┃ ┣ πŸ“‚ paid partnerships ┃ ┣ πŸ“‚ affiliate codes ┃ β”— πŸ“‚ whitelisting ┃ ┣ πŸ“‚ analytics ┃ ┣ πŸ“‚ views vs followers ┃ ┣ πŸ“‚ saves & shares ┃ ┣ πŸ“‚ comments ┃ β”— πŸ“‚ weekly review ┃ ┣ πŸ“‚ content tools ┃ ┣ πŸ“‚ Claude ┃ ┣ πŸ“‚ Canva ┃ ┣ πŸ“‚ CapCut ┃ β”— πŸ“‚ Notion ┃ ┣ πŸ“‚ content wins ┃ ┣ πŸ“‚ hooks that hit ┃ ┣ πŸ“‚ formats to repeat ┃ ┣ πŸ“‚ part 2 requests ┃ β”— πŸ“‚ videos to remake ┃ β”— πŸ“‚ growth ┣ πŸ“‚ followers ┣ πŸ“‚ leads ┣ πŸ“‚ customers ┣ πŸ“‚ revenue β”— πŸ“‚ content scaling save it for the next time you don't know what to post. (the research folder is in the article ↓)
20
16
3
201
15,533
15 design resources worth bookmarking πŸ”– 1. reelfolio.io: animated project showreels 2. inspora.design: creative design inspiration 3. land-book.com: landing page inspiration 4. wonderlist.design: premium Figma designs 5. landdding.com: landing page examples 6. footer.design: footer design ideas 7. cta.gallery: CTA design examples 8. motionin.design: interactive web sections 9. 60fps.design: UI animation references 10. iconly.design: app icon inspiration 11. fontinlogo.com: fonts behind famous logos 12. backgrounds.supply: unique backgrounds 13. grainient.supply: gradients and animated backgrounds 14. on.design: designer community 15. v0.app/templates: ready-made app templates
14
85
1
615
31,160
Hadi retweeted
HOLY F*CK, motion studio in 20 repos. 20 open-source skills for Opus 5.5. Animation. Physics. 3D. Sound. Delivery. 01 Remotion β†’ React video engine github.com/remotion-dev/remo… 02 HyperFrames β†’ HTML-to-video renderer github.com/heygen-com/hyperf… 03 Anime.js β†’ Timeline choreography github.com/juliangarnier/ani… 04 React Spring β†’ Spring animation github.com/pmndrs/react-spri… 05 PixiJS β†’ GPU-powered 2D graphics github.com/pixijs/pixijs 06 p5.js β†’ Generative visuals github.com/processing/p5.js 07 Paper.js β†’ Vector paths github.com/paperjs/paper.js 08 Matter.js β†’ 2D physics github.com/liabru/matter-js 09 Three.js β†’ 3D scenes github.com/mrdoob/three.js 10 Rapier β†’ 3D collisions github.com/dimforge/rapier 11 OGL β†’ Custom shaders github.com/oframe/ogl 12 tsParticles β†’ Particle effects github.com/tsparticles/tspar… 13 glTF Transform β†’ 3D asset optimization github.com/donmccurdy/glTF-T… 14 GL Transitions β†’ Shader transitions github.com/gl-transitions/gl… 15 OBS Studio β†’ Screen capture github.com/obsproject/obs-st… 16 Chatterbox β†’ Voiceovers github.com/resemble-ai/chatt… 17 Tone.js β†’ Synths and sound effects github.com/Tonejs/Tone.js 18 faster-whisper β†’ Caption transcription github.com/SYSTRAN/faster-wh… 19 OpenCut β†’ Timeline editing github.com/OpenCut-app/OpenC… 20 FFmpeg β†’ Encoding and delivery github.com/FFmpeg/FFmpeg Brief β†’ prototype β†’ animate β†’ render β†’ ship. Bookmark this. Go build something worth watching ↓
77
554
14
5,112
304,432
how are you making these ai-generated videos so cheap? i'm trying a 15 sec video with seedance 2.5 at 480p. i tried 4 times and it spent $6. is there a more affordable way to do it?
307
184
11
3,952
738,972
Damn Claude is so good at video editing it's ridiculous
183
64
16
2,634
194,135
Hadi retweeted
Replying to @jordansgoodman
we launched the most comprehensive ai performance engineering repo in the world last week now we'll be posting every single resource this is Wafer's ai performance engineering series save this as your starting point. links in thread 🧡 part 1: "All About Transformer Inference" from How To Scale Your Model. - the authors cover the computations, memory traffic, and serving decisions behind transformer inference: - arithmetic intensity of linear layers and attention across prefill and decode. - kv cache sizing by layer count, kv heads, head dimension, sequence length, and precision. - the compute/hbm bandwidth crossover and how batch size and quantization shift it. - decode latency and throughput bounds from parameter bytes, kv bytes, and hardware bandwidth. - weight reuse through batching and diminishing throughput gains as kv traffic grows. - gqa, kv quantization, and PagedAttention, including the memory costs each addresses. - model sharding, kv placement, and collective communication overhead. - continuous batching, prefix caching, and disaggregated prefill/decode. the worked problems connect model dimensions to deployment decisions like memory capacity, workload distribution, and expected performance under the stated assumptions.
2
51
18,387
India got latent member's only bonus episode 4k
11
109
1
928
140,909
Hadi retweeted
Do this for 1 month and you’d almost not recognize your body anymore
πŸ“·πŸ“·Brutal Core Workout
270
6,948
132
122,703
10,476,283
If you want to become a world-class AI engineer in 2026, watch these 12 YouTube videos (save them):
43
462
9
2,864
388,838
We analyzed every app doing over $1M in MRR to figure out exactly what separates the massive winners from the ones that stall out. We found the pricing models, retention mechanics, and paywall structures behind these apps, then documented the entire framework into an 11-point cheat sheet. We're giving it away for free. If you run a consumer app and want to: > Build a retention engine that actually works > Increase your day-35 conversion rates > Scale a profitable subscription business Comment "MRR" and I'll DM you the full cheat sheet for free (must be following for the DM to send)
835
40
5
721
98,074
CLAUDE GETS YOU MORE INTERVIEWS THAN ANY RESUME TEMPLATE EVER WILL. COPY THESE 6 PROMPTS:οΏΌ
2
33
1
234
30,382
πŸ””BREAKING: Claude AI can now help you build a $1B company strategy. Use this prompt:οΏΌ
6
82
680
66,024
Our calming spray is designed to reduce anxiety and promote deep, restful sleep β€” safely and gently. Less stress. More tail wags. 🐾 πŸ’™ #DogAnxiety #PetSolutions #DogCareTips #PetAnxiety #CalmingSpray #DogLife #FurBaby #PetLovers #DogHealthCare #PetBrand #Shorts
1
90
Our AI UGC model takes over the city wearing next-gen denim jackets β€” bold, effortless, street-certified. From crosswalk moments to late-night city vibes, this is where fashion meets the future. Would you rock this fit in Tokyo? πŸ‘€πŸ‘‡ #GenZFashion #AIugc
1
164