harsh retweeted
I get inspired by literally everything: old ads, packets, signboards.
But Indian vintage design is so hard to find. It’s scattered across old newspapers and random blogs.
So I made Virasat, an online museum of India’s everyday design 🧵
virahst[.]vercel[.]app
harsh retweeted
I get inspired by everything: old ads, packets, signboards.
But Indian vintage design is hard to find. It’s scattered across old newspapers, blogs and random corners of the internet.
So I built Virasat, an online museum of India’s everyday design 🧵
virahst.netlify.app
harsh retweeted
so many books to read, instruments to learn, languages to speak, poems to write, countries to visit, songs to sing, movies to watch, people to meet, pictures to take, food to taste, places to see, sunsets to watch, stories to hear, stories to tell, things to create, hobbies to discover, dreams to chase, love to give, memories to make, lessons to learn, and a whole life to live.
People use Jev to pick a model before a task.
I made it change GPT-6's reasoning effort inside Codex DURING the task.
More thinking when stuck. Less for routine steps.
50% lower Astra costs in my tests. Faster runs, without breaking prompt caching.
harsh retweeted
Let them disqualify you. Dont ever disqualify yourself.
You’re unable to view this Post because this account owner limits who can view their Posts. Learn more
harsh retweeted
Think Quant Memo is just quant finance? Think again.
Algorithmic trading. Machine learning. Deep learning. Neural nets and LLMs. Stochastic calculus. Derivatives. Market making. And the research shaping all of it. One place. All free.
We're here to disrupt, and we're coming straight for the paywalls.
Something new ships every week. Take what they've been charging you for 👉 quantmemo.com
harsh retweeted
Beware people obsessed with outcomes instead of building outcome machines. Its worse than ever with AI, but these people existed before. Short term results above all else, etc. Don't fall into the trap. Invest in building strong fundamentals, invisible supports, and outcomes flow like water. An outcome machine.
harsh retweeted
Stats which made YC give us an interview this time:
(on our 3 month old product)
>$200k cARR
>$20k revenue
closed two F500 global, four more in pipeline
The 10 min call went like bullet and they loved the idea. Too bad could not show enough US traction
Untill next time? 💪
Introducing Lunel
Your entire dev environment, now in your pocket.
No login. No signup.
Just scan the code and start building.
Available on iOS and Android.
Everything you need to ship, in one mobile app:
• AI agents - OpenCode, Codex, Claude Code (soon)
• Use AI agents with just voice
• Full editor
• Built-in browser with DevTools
• Terminal
• Git
• Even Brainrott
… and a lot more.
for more details check reply :)
harsh retweeted
Vibe-coded entire context graph database in ~7 days. Introducing OpenContext
a tool called 'zep' gave my agents best retrieval on context graphs but was v expensive, so took an open source repo and made all the proprietary features of zep on top of it. A drop in replacement to Zep, just change the URL and API_KEY while you self host everything
* top tier retrieval with sub 300ms latency
* admin dashboard to visualise your nodes
* fully open source
* based on research by PhD holders
harsh retweeted
When people say there aren't any jobs for new grads anymore, show them this.. not saying it's fair, but job market has always been winner-take-all, and even more so in the last few years.
harsh retweeted
Presenting learn.sol
LearnSol helps you go from zero to deploying solana apps in one place
- guided tutorials across solana, rust, anchor and client-side
- ai chat that explains any concept with full context and citations
- game based learning where you play, answer and earn nfts on devnet
- practical projects like escrow, token mint, nft marketplace with one click deploy
- coding challenges like 30 days of rust to stay consistent
instead of scattered docs, repos and endless setup you get a single place to learn, practice, build and showcase your skills
simple, structured, no bs
try out our early release here - learnsol.site
harsh retweeted
Replying to @vercel
@vercel @vercel_support have an urgent issue related to bandwidth usage. Want to know if VPC peering will solve it. Can anyone help? Have reached out to support as well. @rauchg
harsh retweeted
Moved our database to @PlanetScale Metal Postgres last night. Early numbers on a heavy API are 🔥:
p50 334ms → 92ms (-72%)
p75 450ms → 126ms (-72%)
p90 500ms → 170ms (-66%)
Massive thanks to the PlanetScale team for help with the migration, and quick hands on support. @mattrobenolt @samlambert
harsh retweeted
"Learn to sell, learn to build, if you can do both, you will be unstoppable."
@naval























