screenpipe (YC S26) retweeted
curently building at @screenpipe x @42born2code
{Soulbook} is gonna be a very successful compagny , or a crazy screenpipe feature😆
@louis030195
My bag is ready i have my laptop and 2 bottles of water just dm me the flight time. see you all in SF🌉
Best sponsor out of the last 3 hackathons tho truly amazing product @screenpipe
screenpipe (YC S26) retweeted
🎖️just won a $1000 prize at the @screenpipe (YCS26) x @42born2code hackathon today in paris 🗼
that was really fun, loads of cool people and a really interesting product, thanks to my team and overall to everyone, that was fucking fire
screenpipe (YC S26) retweeted
Tinfoil helps us be the most private workflow memory product!
Screenpipe is an open-source, local-first app that records how you work and gives you and your agents a long-term memory. Tinfoil helps @screenpipe reach regulated customers w/ privacy
"I don't know of any other alternative that would unblock us here." - Louis Beaumont, Founder
screenpipe (YC S26) retweeted
Introducing @river_ai_inc x Screenpipe integration
Train a model on how people work in your company, or your personal super intelligence
I trained a model on 7 years of notes + Screenpipe data
screenpipe (YC S26) retweeted
i started using using local models (qwen3.5) via @lmstudio and also using @screenpipe for always on local screen recording.
And i'm finally hearing the Apple M4 Max fans running. Forgot what Intel macs used to be like to work with lol.
screenpipe (YC S26) retweeted
“In a world where notes are managed by agents, the app now has zero value.”
Exactly why we built screenpipe differently: open source, mp4 files + transcriptions on disk.
Notes UI is a dead end. The context is the product.
Recordings + transcripts stay local by default, ready for the AI you choose.
xitter.cf/appenz/status/20335973…
Sorry to see Granola @meetgranola going closed. They encrypted their local db, no local and no cloud API. In a world where notes are managed by agents, the app now has zero value.
Any recommendations for good alternatives? What are you switching to?
screenpipe (YC S26) retweeted
I just joined Y Combinator Summer 26. I built 200+ open source projects before (without AI), here's my story:
screenpipe (YC S26) retweeted
GitHub 爆火项目榜:本周新增星标精选 TOP 7。
我最喜欢推荐7 screenpipe 最符合我对 AI 时代个人工具的期待,它不是一次性玩具,而是能持续积累价值的“第二大脑”。
这些项目在本周星标增长明显,涵盖 AI 视频生成、虚拟伴侣、安防监控、物理仿真、照片管理等领域。
1 MoneyPrinterTurbo +7948
【造富潜力】
仓库:github.com/harry0703/MoneyPr…
简介:利用 AI 大模型,一键生成高清短视频。输入主题或关键词即可自动完成文案、素材、字幕、配乐并合成完整视频,支持中英文。
亮点:全自动流水线 + WebUI,适合内容创作者快速生产。
造富潜力:项目名称本身带有“印钞机”意味,社区常将其视为批量生成短视频的工具,适合尝试制作无脸视频或带货内容并上传抖音、YouTube、TikTok 等平台进行流量变现。但实际效果高度依赖视频质量、平台算法、内容合规以及个人运营能力。目前社区讨论多为潜力宣传,尚未有大规模公开验证的成功收益案例。
2 airi +1079
仓库:github.com/moeru-ai/airi
简介:自托管的 AI 虚拟伴侣,支持实时语音聊天、可玩游戏(Minecraft / Factorio 等),兼容 Web / macOS / Windows。
亮点:完全本地部署,可自行控制。
3 Frigate +1308
仓库:github.com/blakeblackshear/f…
简介:实时对象检测的本地 NVR,专为 IP 摄像头设计,集成 Home Assistant,使用本地 AI 完成物体识别。
亮点:隐私优先,无云端依赖,智能家居用户常用。
4 genesis-world +247
仓库:github.com/Genesis-Embodied-…
简介:通用机器人与具身 AI 的仿真平台,集成多物理引擎和照片级渲染器,用于物理 AI 开发。
亮点:生成式世界模拟,适合机器人和强化学习研究。
5 ppf-contact-solver +1603
仓库:github.com/st-tech/ppf-conta…
简介:物理仿真接触求解器,支持布料、实体和杆的复杂物理模拟。
亮点:高性能接触求解,适用于游戏开发、材料仿真等领域。
6 PhotoPrism +68
仓库:github.com/photoprism/photop…
简介:AI 驱动的自托管照片管理应用,类似私有版 Google Photos,支持自动标签、智能搜索和 RAW 处理。
亮点:完全本地运行,隐私优先。
7 screenpipe +146
仓库:github.com/screenpipe/screen…
简介:本地屏幕 + 音频持续录制工具,24/7 捕获数字生活,生成可搜索的 AI 记忆库。
亮点:把电脑变成有记忆的个人 AI 引擎,支持后续个性化工作流。
最近是不是都在被 GPT-Image2 刷屏?
一键漫画、插画、绘本,整体效果确实很赞。
但用多了基本都会遇到一个问题:
👉 细节里的字还是有部分容易出错。
推荐有需要做漫画 / 插画 / 海报收藏住
RefineAnything(图像局部精修)这个刚需级工具。
limuloo.github.io/RefineAnyt…
它的核心就一件事:
只修你选中的地方,其他一动不动。
• 专门针对“文字崩坏 / 小细节错误”优化
• 只改局部,不会把整张图风格改掉
• 修完之后,排版 / 色彩 / 氛围完全保持原样
———
用法附在图上了(全程浏览器):
1️⃣ 把错误的区域随便涂一下
2️⃣ 写一句:refine the text "被褥"(或你要修的字)
3️⃣ 点运行,等 20 秒
基本不用调参数,小白也能直接用。
👉 它专门训练了“文字精修”这件事
所以修出来的字:
结构是对的,笔画是完整的,
而且不会影响原本画风。
screenpipe (YC S26) retweeted
My AI employees:
- ChatGPT - Project manager
- Replit - Developer
- Grok - Market research
- Manus - Secretary
- OpenClaw - COO
- Higgsfield - Video producer
- Gemini - Graphic Designer and Editor (of images and files in GDrive)
- Origami - SDR
- Wispr - Stenographer
- Screenpipe - All-knowing sage
- G.E.N.A. - can’t talk about this yet
Total monthly cost ~$720 (before paying for additional credits as needed)
This isn’t the entire stack I use, but it’s the main ones.
Others I didn’t list—Midjourney, Apollo, Gamma, Siri, Plaud, Canva, Alexa, Claude, Claude Desktop, Codex, Cursor, Google (in search—I use this a decent amount), some of my own AIs that I’ve created using various APIs that tie into my private custom tools
screenpipe (YC S26) retweeted
i just lost a billion dollars
3 years ago Cursor CEO dm'd for a founding engineer role and I ignored him because I wanted to build my own startup
hold me accountable x, I'm gonna make more than a billion with screenpipe
screenpipe (YC S26) retweeted
The world is about to become hyperconnected and cross the percolation threshold.
Thinking about 1:1 AI replacing human work is peak midwittery, classic SF technofaghitree.
AI is about restructuring the sociocybernetics of all civilization and human interaction/connectivity.
screenpipe (YC S26) retweeted
screenpipe PII model live demo
9 ms inference on consumer device, outperform frontier model
screenpipe.github.io/screenl…
screenpipe (YC S26) retweeted
your memory forgets. Screenpipe doesn't.
GitHub Repo of The Day for Creators - Day 8
name : 𝘚𝘤𝘳𝘦𝘦𝘯𝘱𝘪𝘱𝘦
→ what it does
a local AI-powered memory system that continuously records your screen activity and transcribes audio running through your computer.
it automatically:
- reads on-screen text with OCR
- transcribes meetings and conversations
- indexes everything locally
- creates a searchable history of your digital life
think of it as search for your past.
→ why creators care
everyone has experienced this.
you saw an article.
a tweet.
a metric.
a code snippet.
a client message.
then three days later you need it again and have absolutely no idea where you found it.
@screenpipe solves that problem.
instead of relying on bookmarks, screenshots, or memory, you can simply search for a keyword and jump back to the exact moment it appeared on your screen.
→ creators that need it
- founders
- freelancers
- researchers
- developers
- operators
- content creators
- consultants
anyone who constantly switches between tabs, meetings, documents, and tools.
→ what problem it solves
- forgotten client requests
- lost articles and websites
- scattered research
- missed meeting notes
- untracked work sessions
- context switching overload
it creates a searchable record of what you've seen and heard.
→ how creators actually use it
- search old client meetings for exact quotes
- recover websites they forgot to bookmark
- find metrics they saw days ago
- locate code snippets from previous sessions
- generate end-of-day work summaries automatically
- connect local AI models to analyze work history
→ why this repo stands out
most tools that promise "perfect memory" come with privacy tradeoffs.
Screenpipe takes a different approach.
it's:
- open source
- local-first
- private
- self-hostable
- searchable
- AI-ready
your data never has to leave your computer.
for many people, it replaces:
- note-taking apps
- meeting transcription services
- memory tools
- expensive productivity subscriptions
→ best use case
perfect for:
- founders drowning in context switching
- freelancers managing multiple clients
- developers researching across dozens of tabs
- operators generating automated work reports
- creators collecting ideas and references
→ founder
@louis030195
→ extra proof
- actively maintained
- growing plugin ecosystem ("Pipes")
- recent releases and commits in 2026
- community-built automations already connecting Screenpipe to tools like Notion, Salesforce, and HubSpot
this is one of those repos that feels unnecessary until you use it.
then you start wondering how you ever worked without it.
this repo has 80k stars on Github for a reason, here's why.
GitHub Repo of The Day - Day 7
name : 𝘓𝘰𝘤𝘢𝘭𝘚𝘦𝘯𝘥
→ what it does
a free and open-source app that lets you instantly send files, folders, photos, videos, and text between nearby devices.
Windows.
macOS.
Linux.
Android.
iPhone.
iPad.
everything works together.
think AirDrop, but for every device you own.
→ why creators care
moving files between devices should not be a project.
yet people still:
- upload to Google Drive
- email files to themselves
- use WhatsApp and lose quality
- wait for WeTransfer uploads
- fight Bluetooth connections
LocalSend skips all of that.
pick a file.
select a device.
send.
done.
→ creators that need it
- video editors
- photographers
- content creators
- freelancers
- students
- marketers
- remote workers
if you regularly move files between devices, you'll probably use this every week.
→ what problem it solves
- cloud upload delays
- file size limits
- quality loss from messaging apps
- privacy concerns
- account signups
- platform lock-in
everything stays on your local network.
nothing needs to touch the cloud.
→ how creators actually use it
- transfer camera footage from phone to laptop instantly
- send client contracts during meetings
- move project assets between editing devices
- share lecture recordings with classmates
- transfer family photos and videos without compression
it feels like magic the first time you use it.
→ why this repo stands out
most file-sharing tools force you into a cloud workflow.
LocalSend doesn't.
it's:
- completely free
- open source
- offline-first
- encrypted
- unlimited file sizes
- cross-platform
you get AirDrop-level convenience without being locked into a specific ecosystem.
→ money saved
eliminates the need for many paid file-sharing and cloud storage workflows.
zero subscriptions.
zero ongoing costs.
→ best use case
perfect for:
- creators moving media between devices
- editors transferring project assets
- freelancers sharing files with clients
- students exchanging large datasets
- families sharing photos and videos privately
→ extra proof
- 80k+ GitHub stars
- thousands of forks
- active community
- regular updates
- one of the fastest-growing utility repos on GitHub
this is one of those tools you install once and immediately wonder why every operating system doesn't include it by default.
screenpipe (YC S26) retweeted
AI has no idea what people are doing in your company ... unless your company is powered by @screenpipe
Imagine replacing 90% of your employees with a team of geniuses who have no idea how your company operates.
Total chaos. Nothing works.
That’s what AI feels like today.
The missing piece is extracting all the domain knowledge from people’s heads and providing that as structured context to the models.
screenpipe (YC S26) retweeted
one week after getting into YC S26 @screenpipe weekly active users more than tripled 🥲
A year from now, working without giving AI access to everything you do all day will look insane.
Every company will ingest what every employee does, 24/7, into the brain of their AI.
screenpipe (YC S26) retweeted
I'm looking for an AI Phd intern to work on a new kind of AI model at @screenpipe
DM open :)
screenpipe (YC S26) retweeted
one of the reason I built @screenpipe is because I'm a heavy @obsdmd user - I have a personal second brain that captures everything I've thought, written, decided for years. But Obsidian only knows what you type into it
screenpipe knows everything you've seen, said, or heard - automatically
And when you point that at a company instead of a person, you get the institutional second brain: every workflow, every decision, every departing employee's tacit knowledge - captured, queryable, and convertible into SOPs and agents
screenpipe (YC S26) retweeted
When I applied to YC a few years ago for the first time I had built a GPT2 powered Discord bot that was animating communities with thousands of people 🥲
I was running GPT2 when it came out with RAG powered by a FAISS vector search index with thousands of rows I had labelled manually through a UI I built to do some kind of RLHF
I had no business notion though 🥲 pure yolo mode
































