@BoazWith@kwangwsim@opencutapp@Cloudflare Both. Delete project → nukes that project's OPFS media + IndexedDB in one shot. Remove a clip → same immediate cleanup. Import checks quota with 50MB headroom. Beyond that we lean on browser storage pressure. Gap: no LRU, no cross-project dedup, no storage dashboard yet.
@kwangwsim and I rewrote @OpenCutApp to run fully on @Cloudflare using Vite + React instead of Next.js.
Now it’s fully local-first with IndexedDB + browser-native video APIs.
Zero backend infra.
https://t.co/4WJnCtIkL2
@BoazWith@kwangwsim@opencutapp@Cloudflare Neither was the bottleneck — because we avoided both becoming one. OPFS stores media, not IndexedDB. Frame extraction taps the browser's hardware decoder via HTMLVideoElement + OffscreenCanvas, zero transcoding. The whole trick was working with browser primitives vs against
@googlegemma we built Stratos Office on this 🙌 Gemma 4 E2B by @GoogleDeepMind powering 30+ local AI tasks via Transformers.js + WebGPU — OCR, transcription, contracts, code review. Zero servers, zero API keys, total privacy #gemma4
@GigaHierz@minipay@Celo wow you are so fortunate to get to travel the world an do what you love..on a side note, think you and @Drewmutable should connect and talk about celo/minipay intergration on @pinatacloud agents
Shipping MiniPay AI Pay-Go on @Celo — pay micro-cents in cUSD to run AI tasks (web search, scraping, people lookup, social data) directly from @MiniPay. No subscriptions. Pay only for real results. Powered by @agentcashdev x402 APIs + on-chain escrow with auto fee splitting. 👇
This is finally happening! Have been working on this for some time and so excited to share and for a lot more to come this year!
@MiniPay is growing a lot in SEA and I want to meet all the amazing miniapp builders!
Please share all contacts and apps with me 🥰🇻🇳
@kwangwsim and I built a TurboQuant fork for Gemma 4 on WebGPU in the browser.
We wanted long-context, local-first inference without server round trips. thank you @xenovacom https://t.co/J1h9nTGTY7
@kwangwsim and I built a TurboQuant fork for Gemma 4 on WebGPU in the browser.
We wanted long-context, local-first inference without server round trips. thank you @xenovacom https://t.co/J1h9nTGTY7
Built an OpenAI-compatible LLM gateway on @Cloudflare Workers: chat completion
s, embeddings, model discovery, and access to text, image, and speech models
through one simple API. https://t.co/UXIpvGqCIh
Great to see Da Nang turning into a real Web3 hub 🇻🇳
More builders, nomads, and IRL community coming together.
Shoutout @sqrDAO@LiskSEA for helping make it happen.
EMpower Founders Residency just opened — 15 SEA founders in Da Nang this May 🧵
New wave of “agentic video editors” being built with @claudeai Code + @Remotion + @fal@replicate@togethercompute@Kling_ai@runwayml
Generation is the easy part.
The real difficulty is the pipeline: uploads, ASR on long videos, semantic search, rendering and storage.
I built an app where you upload your raw media and it turns it into a smooth narrated video, with clips aligned to the script
no prompts, no clip arranging, no voiceover recording, no music selection - everything is done automatically
the best part: after a late night event, when the last thing you want to do is edit videos, Reelful turns your photos and clips into a recap in minutes!
Just open sourced Bhutanese Duolingo, a Dzongkha learning app with script study, beginner
lessons, flashcards, spaced repetition, and games.
Live: https://t.co/5GcnUWysiB
Code: https://t.co/QTxnpidkGS
#Dzongkha#OpenSource#LanguageLearning#Bhutan
Tibetan Duolingo for learning Tibetan script with alphabet study, flashcards, spaced repetition, typing drills, and games.
https://t.co/jpOWckWMbO
Pleae feel to borrow, improve, and build more from it.
Code: https://t.co/EeHjoNeVdM
#OpenSource#Tibetan#learntibetan
@_tenZdhon_ Ya but codex has so much better usage, tell Claude to make the plan, let codex evaluate and provide comments to plan, let Claude update accordingly, let codex build let Claude audit the execution