Marketing | Systems | Growth | Books | some Digital #Marketing Nerd | All about 🍔🍟 Views my own! ✌#Ifollowback #politicallyincorrect
Riyadh
Joined May 2011
- Tweets18.8K
- Following1.5K
- Followers512
- Likes11.1K
Don retweeted
holy f*ck, if you’re still paying a big LLM to make tiny decisions in 2026, you need to see this before 2027
cut your Claude Code or any LLM bill, make your AI agents safer and run AI over a million rows of data for pocket change.
and the craziest part? the model behind all this can’t even write a sentence.
it’s called Jev.
it only makes quick decisions: which file, is this safe, which category, yes or no. input costs $0.042 per million tokens, output is free.
so why keep paying a big LLM to do the easy stuff?
FOR YOUR CODING AGENT:
-> jevgrep: Claude stops reading half your repo
you ask what the code does, it finds the right files. same tasks solved, about 30% cheaper
github.com/dzhng/jevgrep
-> jev-guard: a bodyguard for Claude Code, Codex and Cursor
checks every command before it runs. rm -rf gets blocked, a sketchy web page gets flagged. about 4 cents per 1,000 checks
github.com/leepokai/jev-guar…
-> leash: your CLAUDE.md rules actually get followed
"no TODOs", "no lazy abstractions". it checks every turn in about 300 ms and makes the agent fix it
github.com/CMaintz/leash
FOR YOUR DATA:
-> duckdb-jev: AI on every row, straight from SQL
2,049 rows classified in under 1 second. the secret is batching: one row at a time ran at 6 rows/s, batched it hit 1,943 rows/s
github.com/prasanthj/duckdb-…
-> jevframe: AI inside pandas and Polars
label reviews, sort support tickets, flag urgent messages. results come back as a normal DataFrame column
github.com/ktaletsk/jevframe
-> pg-jev: Postgres that understands English
SELECT * FROM tickets WHERE jev(tickets, 'the customer is angry'). no embeddings, no vector database
github.com/realZachi/pg-jev
-> jev-calibrate: test before you trust it
the first draft of one question got only 18 of 26 labeled examples right. this tunes it on your own data until it's reliable
github.com/smkrv/jev-calibra…
-> Ollaya: like Ollama, but for decision models
run Jev-style models on your own machine, no API bill, and plug them into Claude Code as an MCP
github.com/ollaya-dev/ollaya
the takeaway is stupidly simple:
-> let small models handle the easy decisions
-> save your big LLM for the stuff that actually needs intelligence
-> batch your data instead of wasting money row by row
-> test everything on your own labels before trusting it
the developers who figure this out first won’t necessarily have better AI models. they’ll just stop wasting money on the wrong ones.
want the prompt for this Jev dashboard? comment “JEV” or DM me and I’ll send it over.
Don retweeted
today, we introduce a little mischief to the world.
for four years, we’ve been brewing real brain cells on electrodes to understand how they compute. now, we’ve turned those discoveries into a video model anyone can use.
from 0.5¢ a second, mischief 1 is at least 5x faster than its base and the cheapest image-to-video model on @fal and @OpenRouter. soon to be on @awscloud marketplace.
it’s not a delusion anymore. biocompute is real, commercial and this is just the beginning.
happy halloween.
Don retweeted
Open Sourcing Higgsfield Canvas!!
Run your own models locally through ComfyUI. No subscription, no ads. MIT licensed. Feel free to build your own canvas
Built to be pluggable: bring any AI-video provider’s API.
repo: github.com/raph651/open-sour…
#OpenSourceHiggsfield #Comfy #AIGC
Don retweeted
R.I.P. Higgsfield Katana.
We reverse engineered it into a Claude skill you can feed your agent.
Download the skill: scenario.com/explorations/ki…
> describe your product, game or brand in a few lines
> Claude writes, animates and cuts the whole film to the beat, 4K out
> Scenario makes the music and sound effects
Higgsfield influencers kept promising the Katana setup and never sent it, so here's the whole skill.
Don retweeted
구글 가족계정으로 제미나이 사용량 6배 만들기!🚀
구글 가족계정을 보다가 문득 든 생각..
이거 가족들도 다 똑같이 혜택 받는데,
내가 내 가족이 되면..? (총 여섯명이 공유가능)
계산하면 정가 29,000원 x 6 = 174,000원 어치를
AI Pro 학생요금으로 7,300원에 쓰는게 가능해짐..
많이들 알고 계시겠지만 모르시는 분들 위해
어디까지 되고 어디까지 안되는지 알려드림!
일단 이미 여러개 갖고있던 구글 계정을
테스트를 위해 가족으로 모두 묶었습니다!
총 다섯명까지 내 가족 계정으로 초대할 수 있으니까
일단 총 다섯개의 다른 계정을 가족그룹에 초대합니다.
AI Pro 혜택 받을 수 있는 페이지를 모두 들어가봄.
결론부터 말씀드리면..! 생각보다 더 미쳤음😱
(별도사용량 ✅, 통합사용량♻️)
- ✅ 제미나이 사용량 x 6 (계정별로 별도 적용)
- ✅ 노트북LM 노트북수, 소스수, 사용량 x 6 (제미나이와 별도, 계정별로도 별도)
- ✅ 구글 플로우 매월 1050크레딧 x 6 (모든 계정에 지급)
- ✅ Jules (비동기 깃헙레포 코딩 에이전트) 월 100세션 (계정별로 별도 적용)
- ♻️구글 클라우드 콘솔 매달 10달러 크레딧은 대표계정만
- ✅ 구글 클라우드 콘솔 가입시 주는 3달간 300달러 무료 크레딧은 각 계정 별도 적용 (3달에 하나씩 가입하기)
- ✅ 구글 클라우드 콘솔의 크레딧으로 vertex api 받아서 구글 AI 모델들 api로 사용 가능. (Vertex가 에이전트플랫폼으로 이름바꿈)
- ♻️ 안티그래비티 사용량은 전체 계정 통합 사용량
- ♻️ 구글 드라이브 5TB (pro 기준) 전체 계정 통합
- 🤔 구글 포토 각 계정 15GB 이상 용량은 통합
어쨌든 일회성 클라우드 크레딧으로 한화 40만원 x 6
240만원 어치 구글 api 토큰 사용량 +
매달 29000원 짜리 제미나이 사용량 &
노트북LM 사용량 6계정 팡팡 쓰기 +
구글 플로우도 1000크레딧 구매하려면 한화로
약 15,000원 어치임 그걸 x 6 하면 매달 약 7만원! +
그리고 사실 클라우드 5테라도 통합이지만 차고 넘침.
거기다 안티그래비티 사용량과 구글 포토 사용량까지..
일회성 1800달러 + 매달 최소 100달러 이상의 가치임!
이 정도면 구글 거의 자선사업 단체가 아닌가 싶음
이걸 모를때 한 자리 동생한테 줘버린게 너무 아쉽다 ㅠ
빨리 구글 계정만들고 세팅하세요!!
(만약 결제를 5계정하면 30계정이 되지만.. 거기까진 귀찮)
Don retweeted
this is absolute f*cking free
Obsidian's founders, Shida Li and Erica Xu, put the whole toolkit behind the app on GitHub:
the Web Clipper, the importer, the canvas format and the plugin kit. 7 repos, 42,437 stars
the order that takes you from Obsidian user to plugin author:
1. obsidian-importer
bring notes over from Apple Notes, Notion, Evernote and more as plain Markdown
↳ github.com/obsidianmd/obsidi…
2. obsidian-clipper
highlight and capture any web page straight into your vault
↳ github.com/obsidianmd/obsidi…
3. obsidian-help
the full help docs, written in Markdown
↳ github.com/obsidianmd/obsidi…
4. jsoncanvas
the open file format behind Canvas, so other apps can read your boards
↳ github.com/obsidianmd/jsonca…
5. obsidian-sample-plugin
the template a new community plugin starts from
↳ github.com/obsidianmd/obsidi…
6. obsidian-api
type definitions for the whole Obsidian API
↳ github.com/obsidianmd/obsidi…
7. obsidian-releases
the list every community plugin and theme gets published through
↳ github.com/obsidianmd/obsidi…
steps 5 to 7 turn a weekend idea into a plugin every Obsidian user can install, and a plugin with real users is the portfolio line that gets a developer hired
my own stack, with Obsidian at the core, is in the article below ↓
Don retweeted
These are my favorite GTM engineering accounts on X.
The selection process was rigorous: I like to read their posts, and I thought, “Hey, that’s useful. This is a great account.”
No scoring model.
No industry panel.
Grouped by what I’d follow them for. Plenty of crossover, obviously.
GTM SYSTEMS & AI BUILDS
@retttx — Helped build Clay’s GTM engineering function. Shares how the systems fit together.
@BenyaminHolley — Hands-on builds with Claude Code and lessons from shipping real GTM systems.
@gtm_engineer — Workflow breakdowns, operator interviews, and practical GTM engineering tutorials.
@jai__toor — Builds GTM developer tools at Deepline and shares recipes for acting on technical signals.
@idogoldd — Builds AI GTM agents at Swan and shares lessons from putting agents into production.
GTM STRATEGY & FRAMEWORKS
@brendanjshort — Breaks down how AI-native revenue teams work through The Signal.
@schoeny — An analyst’s perspective on AI, GTM tools, and where the market is heading.
@fivosaresti — Maps the full GTM system, from ICP analysis and market mapping to outreach.
@dan__rosenthal — Detailed frameworks for GTM motions, signal architecture, and AI-powered operations.
@nathanclark — Perspectives on GTM operations, revenue growth, and applying AI to everyday work.
SIGNALS & TARGETING
@ENowoslawski — Deep outbound expertise, from enrichment and buying signals to campaign execution.
@itsalexvacca — Clear breakdowns of buying signals and coordinated GTM execution.
@FlorinTatulea — Practical builds for account scoring, prospecting, and win-back campaigns.
@Nickkonsta — Signal-based outbound and finding the right moment to reach a buyer.
OUTBOUND EXECUTION & INFRASTRUCTURE
@coldemailchris — The daily mechanics of outbound: sourcing, enrichment, campaigns, and follow-up.
@bretonwill — List quality, campaign QA, and outbound systems that hold up at scale.
@AaronxShepherd — Shows how offers, targeting, infrastructure, and execution connect to pipeline.
@RCIllingworth — Cold email infrastructure, deliverability, and consistent execution.
@ankit_saas — Workflows connecting lead sourcing, research, messaging, and automation.
@draprints — LinkedIn and cold email playbooks, from setup to reply handling.
GTM PRACTICE & CAREER
@itsMoveek — Product marketing and GTM perspectives, plus resources for getting into GTM engineering.
@MitchellKeller_ — Practical outbound ideas, AI workflows, and an active voice in the GTM conversation.
@Jesusvivas11 — Shares perspectives on building GTM systems for B2B companies.
SALES CRAFT & PIPELINE
@morganjingram — Outbound systems, sales craft, and keeping the human connection in prospecting.
@TechSalesGuy — Sales storytelling and practical advice on outbound, partnerships, and building pipeline.
Who else should be in my feed?
Reply with other accounts that you think should be on this list.
Don retweeted
one year at Stanford costs more than $60,000
the CEO of Obsidian put his second brain setup on GitHub for $0
Steph Ango shared his vault, his theme and his agent skills: 7 repos, 77,106 stars. the vault template alone is a f*cking masterclass
• layer 1: the vault
> kepano-obsidian - his personal vault template, organized bottom-up
↳ github.com/kepano/kepano-obs…
> 40-questions - what he asks himself at the end of every year and every decade
↳ github.com/kepano/40-questio…
• layer 2: getting the web in
> defuddle - any web page as clean Markdown, built for the Obsidian Web Clipper
↳ github.com/kepano/defuddle
> clipper-templates - ready Web Clipper templates for popular sites
↳ github.com/kepano/clipper-te…
• layer 3: handing it to an agent
> obsidian-skills - teaches Claude Code and Codex the Obsidian CLI and its open formats
↳ github.com/kepano/obsidian-s…
• layer 4: making it yours
> obsidian-minimal - the distraction-free theme, tunable to the last detail
↳ github.com/kepano/obsidian-m…
> flexoki - an inky color scheme for prose and code
↳ github.com/kepano/flexoki
copy layer 1 before layer 3. an agent is only as useful as the vault you point it at
Karpathy's take on the same idea, Claude skills + Obsidian, is in the article below ↓
Don retweeted
Get paid $8,500/month to build AI.
Generator Residency is open:
• $8,500 per month stipend
• Travel covered
• 12 weeks in Berkeley
• Jan 18 – Apr 9, 2027
Applications close Nov 1.
That's 4 days left. Don't scroll past this.
generatorresidency.org
Don retweeted
This is f*cking insane. Andrej Karpathy spent 8 years at OpenAI and Tesla
Last week, he put everything he learned into one free 2-hour lecture
Agents → Harness → Loops → Graphs
Some bootcamps charge $15K and teach you less than this
You might not have 2 hours right now
Bookmark it and watch before it gets buried in your feed
Then read the guide below and build your first agent harness ↓
Don retweeted
I open sourced Higgsfield!
With a goal in mind, make it as easy as possible for anyone to create, in the comfy cloud or on your local machine!
github.com/Trenchplanet/mogf…
Will continue to work on this github, to make it really the best all in one solution for new or experienced creators.
Some features:
- My own take on genjutsu
- ReFrame
- ReStyle
- Character Builder (automatically creates a character board when you upload an image of a character)
- Agent mode (Talk with an agent and make storyboards, scripts and generate multiple shots at once to speed up workflow)
Thanks to @ComfyUI for this challenge!
#OpenSourceHiggsfield #Mogfield
Don retweeted
🚨 EL PEOR ESCENARIO PARA LAS APPS DE IA ACABA DE HACERSE REALIDAD.
Y casi nadie se ha dado cuenta.
Sunra acaba de lanzar una API que junta las mejores IAs de imagen, vídeo, audio y 3D en un solo sitio.
Seedance 2.5. Kling 4.0. GPT Image 2.5. Y muchas más.
Todas con una sola key.
Hasta hoy funcionaba así:
→ Una cuenta en cada laboratorio
→ Una suscripción por cada app
→ Y pagar aunque no usaras nada
Sunra rompe eso.
Cada modelo cuesta exactamente lo que cobra quien lo creó. Sin recargo. Sin suscripción.
Básicamente: todos los laboratorios de IA detrás de una sola puerta… sin peaje por entrar.
Y el modelo que pides es exactamente el que se ejecuta. Cada cargo, desglosado en tu consola.
La guerra de la IA ya no va de quién tiene el mejor modelo.
Va de quién te da acceso a todos.
Don retweeted
GITHUB IS ACTUALLY INSANE RIGHT NOW.
You can build an entire AI content factory without hiring a team.
Here are 10 open-source repos worth checking out:
1. OpenCreator
Give Codex a creator workspace for writing scripts, generating images, creating voiceovers, translating videos, and editing content.
github.com/krillinai/Open…
2. Social Media Research Skills
Use Claude Code or Codex to research viral posts, analyze competitors, mine comments, and extract hooks from video transcripts.
github.com/ScrapeCreators…
3. Local Deep Research
Turn any topic into a detailed research report using web searches, your documents, and local or cloud AI models.
github.com/LearningCircui…
4. FLUX.2
Generate and edit images for thumbnails, social posts, and marketing campaigns. The Klein 4B models have Apache-licensed weights.
github.com/black-forest-l…
5. Kokoro
Turn written scripts into AI voiceovers with a compact, open-weight text-to-speech model.
github.com/hexgrad/kokoro
6. ACE-Step 1.5
Generate music from text prompts, create instrumentals, and regenerate specific audio sections. It can run locally on supported hardware.
github.com/ace-step/ACE-S…
7. Short Video Factory
Create batches of short videos from prompts and your footage, with AI-generated scripts, voiceovers, editing, and subtitles.
github.com/YILS-LIN/short…
8. AI YouTube Shorts Generator
Convert long YouTube videos into vertical Shorts with AI-selected highlights, transcripts, and automatic cropping.
github.com/Anil-matcha/AI…
9. Podcastfy
Turn source material into AI-generated podcast conversations, customize the dialogue, and produce audio in multiple languages.
github.com/souzatharsis/p…
10. Social Media Agent
Turn a URL into X and LinkedIn posts, review and edit the drafts, then schedule them through an AI agent workflow.
github.com/langchain-ai/s…
Research, writing, visuals, voiceovers, music, video editing, and distribution.
You can automate almost every stage of content creation with the right tools.
Which part of your workflow would you automate first?
Want to build your own AI content factory? 👀
Follow @rakib_hossen_ai for more AI tools, open-source gems, and workflows that help you create smarter and faster. 🚀
Don retweeted
AI動画が『それっぽいだけ』で終わるのは、映画の手法を名前で指定していないから。
構図→カメラ→照明→文字→音の順に、この6サイトで言葉を集めると変わる。
① eyecannndy.com
撮影手法の名前から作例を探せる
② aicameramovements.com
カメラワーク46種のプロンプト
③ melies.co/cinematic-techniqu…
撮影テクニック424種。作例とプロンプト付き
④ grokfilm.app
照明・編集・音響まで300の技法を解説
⑤ fontsinmovies.com
映画で使われた書体を作品ごとに見られる
⑥ mixkit.co/free-sound-effects
場面転換の効果音42種など
映画好きは時間が溶けるので閲覧注意。
Don retweeted
i'm begging you to paste this into Claude Code before your next session.
it scans your whole setup and seals the 5 leaks silently eating your Opus 5.5 tokens, with zero hit to output quality.
all 10 leaks are explained in the article.
grab the prompt before your next bill grabs you:
