ben hylak's banner

ben hylak

@benhylak • 51,438 subscribers

we are hiring. cto @raindrop_ai, prev @apple design, @spacex eng

Shorts

introducing killedbyopenai dot com a digital graveyard for everything openai has killed. what am i missing?

introducing killedbyopenai dot com a digital graveyard for everything openai has killed. what am i missing?

372,733 次观看

make music physical again

make music physical again

754,007 次观看

Videos

Anya Rossi

sweetdream.ai

SweetDream.ai•Sponsored•Livecam

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

Exclusive private shows

1.2k viewers online

Private Show

Join now for exclusive access

Free preview available • Premium content

we built the first sane way to debug your agent locally. you can see your traces. codex/claude code can too. this lets them write evals and test your agents automatically. best part: it's completely free and open source. install with 1 line. (github below)

we built the first sane way to debug your agent locally. you can see your traces. codex/claude code can too. this lets them write evals and test your agents automatically. best part: it's completely free and open source. install with 1 line. (github below)

193,654 次观看 • 2 个月前

now, your agent can fix itself. introducing raindrop triage. an agent for finding and investigating agent issues.

now, your agent can fix itself. introducing raindrop triage. an agent for finding and investigating agent issues.

210,551 次观看 • 2 个月前

today, we’re excited to announce raindrop 2.0: self-healing agents. we now train custom models that autonomously detect hidden issues with your agent. i could tell you all about it, but wouldn’t you rather hear it from someone… else?

today, we’re excited to announce raindrop 2.0: self-healing agents. we now train custom models that autonomously detect hidden issues with your agent. i could tell you all about it, but wouldn’t you rather hear it from someone… else?

130,019 次观看 • 1 个月前

on saturday, i vibe coded a little app for collecting + virtually polishing rocks. i just realized today that it works on any object. behold, a polished can of lacroix.

on saturday, i vibe coded a little app for collecting + virtually polishing rocks. i just realized today that it works on any object. behold, a polished can of lacroix.

313,715 次观看 • 3 个月前

don't scroll jack. the users scroll should always map 1:1 with the content moving on the page. otherwise it feels like going from a highway to driving through mud. you can still do very cool transitions without jacking scroll (e.g. below)

don't scroll jack. the users scroll should always map 1:1 with the content moving on the page. otherwise it feels like going from a highway to driving through mud. you can still do very cool transitions without jacking scroll (e.g. below)

35,588 次观看 • 12 天前

earlier this year, we launched the first agent monitoring platform now, we are used by some of the fastest growing ai companies (@replit, Framer, Speak, Clay) and just raised a $15m seed round

earlier this year, we launched the first agent monitoring platform now, we are used by some of the fastest growing ai companies (@replit, Framer, Speak, Clay) and just raised a $15m seed round

418,478 次观看 • 7 个月前

just tried it and... it's really, really bad. leaves me wondering if the demo was fake?

just tried it and... it's really, really bad. leaves me wondering if the demo was fake?

891,729 次观看 • 2 年前

we’re excited to announce trajectory explorer: the first sane way to navigate agent traces. every decision your agent made is now searchable in seconds only in Raindrop

we’re excited to announce trajectory explorer: the first sane way to navigate agent traces. every decision your agent made is now searchable in seconds only in Raindrop

180,265 次观看 • 5 个月前

yesterday, i built the code review app i've been begging twitter for. - live streams change summaries as claude code works - quizzes me about changes before i commit them. - finds bugs locally -- without having to push to GitHub

yesterday, i built the code review app i've been begging twitter for. - live streams change summaries as claude code works - quizzes me about changes before i commit them. - finds bugs locally -- without having to push to GitHub

180,987 次观看 • 5 个月前

today, we're announcing our fundrai-- jk. we're shipping. meet raindrop experiments 🧪

today, we're announcing our fundrai-- jk. we're shipping. meet raindrop experiments 🧪

151,932 次观看 • 9 个月前

Today, we're excited to announce Deep Search. It’s like Deep Research, but for your production AI data. Search for anything, and Raindrop automatically trains little models to accurately classify any topic or issue, across millions of events.

Today, we're excited to announce Deep Search. It’s like Deep Research, but for your production AI data. Search for anything, and Raindrop automatically trains little models to accurately classify any topic or issue, across millions of events.

161,416 次观看 • 1 年前

today, we're introducing self diagnostics: the first ever way for agents to proactively self-report issues they encounter. welcome to the future of agent observability.

today, we're introducing self diagnostics: the first ever way for agents to proactively self-report issues they encounter. welcome to the future of agent observability.

62,959 次观看 • 4 个月前

.Speak is an AI language learning app that serves over 15 million users. cto/co-founder Andrew Hsu discusses how speak approaches agent engineering, the importance of self-healing loops, and how they use Raindrop

.Speak is an AI language learning app that serves over 15 million users. cto/co-founder Andrew Hsu discusses how speak approaches agent engineering, the importance of self-healing loops, and how they use Raindrop

28,375 次观看 • 2 个月前

AI products fail constantly—in ways both hilarious and terrifying. Regular software throws exceptions. But AI products fail silently. Meet Raindrop : the first Sentry-like monitoring platform for AI products.

AI products fail constantly—in ways both hilarious and terrifying. Regular software throws exceptions. But AI products fail silently. Meet Raindrop : the first Sentry-like monitoring platform for AI products.

140,621 次观看 • 1 年前

who said b2b saas has to be ugly? new in Raindrop: conversation x-rays🩻

who said b2b saas has to be ugly? new in Raindrop: conversation x-rays🩻

49,417 次观看 • 5 个月前

AI companies today are faced with a choice: build great AI products, or protect user privacy. That compromise ends today. Introducing Raindrop’s PII Guard.

AI companies today are faced with a choice: build great AI products, or protect user privacy. That compromise ends today. Introducing Raindrop’s PII Guard.

78,397 次观看 • 1 年前

took a break from b2b saas to build an ai native terminal. most of my workflow is just pressing enter now.

took a break from b2b saas to build an ai native terminal. most of my workflow is just pressing enter now.

158,063 次观看 • 2 年前

in the future, ai will improve the products you use, before you even have a chance to leave feedback. to that end, i’m excited to announce Dawn, which i’ve been building with my best friends zubin koticha and alexis

in the future, ai will improve the products you use, before you even have a chance to leave feedback. to that end, i’m excited to announce Dawn, which i’ve been building with my best friends zubin koticha and alexis

117,143 次观看 • 2 年前

today we're launching Dawn -- the first product analytics platform built for ai products. with daily slack digests, automatic topic detection, powerful filtering, and semantic redaction rules, we're making it easier than ever to keep a pulse on your ai product.

today we're launching Dawn -- the first product analytics platform built for ai products. with daily slack digests, automatic topic detection, powerful filtering, and semantic redaction rules, we're making it easier than ever to keep a pulse on your ai product.

96,035 次观看 • 1 年前

at Raindrop, we like to treat every prompt as a function: structured inputs, structured outputs every function used to prepare the input for the model lives in the same prompt file, and we have an extension for iterating on the prompt right inside our codebase.

at Raindrop, we like to treat every prompt as a function: structured inputs, structured outputs every function used to prepare the input for the model lives in the same prompt file, and we have an extension for iterating on the prompt right inside our codebase.

31,575 次观看 • 7 个月前