tech media is boring but not me | building cool stuff | DM for work & Collabs . mail: [email protected]
Joined May 2024
- Tweets2.3K
- Following4.5K
- Followers10.1K
- Likes4.9K
Priya retweeted
Most AI video tools stop at making the video.
Fastlane v2 goes further. It creates an AI creator for each audience, makes their content and posts it across IG, TikTok and Shorts.
Real-looking content at this scale is going to change what "UGC" even means.
🤝 Paid partnership
Priya retweeted
Another model upgrade, but this one is worth watching.
Ling-3.1-flash-1019 hits 1524 Elo on AA-Briefcase and 68% on AutomationBench.
AI models are getting better at handling complex tasks, not just generating text.
Open weights are coming. Things could get interesting.
Priya retweeted
holy shit, they just killed gold.
Last month, we demonstrated high-rate alchemy: turning mercury into real, stable gold, beating CERN’s record by a factor of 10.
Last week, we enriched isotopes in the world’s most powerful plasma centrifuge, beating the Soviet record that stood for 40 years.
Now, we’re building a fusion reactor in San Francisco to achieve record power density.
Priya retweeted
marketing in 2026 is going to get absolutely insane.
Higgsfield Content Factory lets a solo founder create more ad variations than an entire marketing team could dream up.
we’re still so early.
🤝 Paid partnership
Priya retweeted
Chinese AI labs watching a model outperform larger models on coding benchmarks while casually running at 125 tokens/sec on a MacBook Pro:
🤝 Paid partnership
Introducing Flint, @Recursive_SI’s first model: a fast local coding model built with our automated research system, the Eureka Machine.
• Best-in-class fast local coding
• Up to 2x faster than comparable models: 125+ tokens/sec on a MacBook Pro, 150+ on RTX Spark
• Built from scratch in under 4 months
Private, unmetered intelligence on your own machine. Open weights available now on @huggingface, run it locally on your favorite framework or try it for free on @opencode or @pidotdev.
Priya retweeted
holy shit, AGI might be a lot closer than people think.
AI is now breaking software on purpose, finding bugs and fixing them before a single user touches it.
1/ Today we're introducing Breken, and with it the final job in software: production engineering.
Your tests check what your team thought of. Production checks everything else. Today, your users run that second test for you.
Breken runs it instead. Your real software, in worlds that behave like production, down to the chaos and the users, finding and fixing what breaks before a single user does.
Here it is: breken.ai
Priya retweeted
Holy shit, an ex-Stripe and Uber engineer just solved a problem every company building AI agents is going to run into.
falconer keeps company knowledge and documentation in sync with the code, so agents aren’t making decisions based on outdated information.
the next big bottleneck for AI agents is context, not intelligence.
🤝 Paid partnership
Trillions are pouring into the race to build the first general superintelligence, but leveraging your own specialized know-it-all is now possible as of today.
I built the knowledge systems at @stripe and @uber. Then I helped Anthropic, Metronome, and Graphite build a killer developer experience. Everyone wants Stripe-level quality, but that bar is too low.
Falconer gives you and your customers complete, specialized intelligence: an internal knowledge base and external documentation in one beautiful package. Falconer is the ultimate trust engine to manage your context: self-driving, observable, and agent-friendly.
I co-founded @falconer_ai with @mbenedetto212 and assembled a team of know-it-all builders. We're backed by fellow know-it-alls @thegp, @BoxGroup, @svangel, @karimatiyeh, @benln, @zachlloydtweets, @CeciStalls, @MerrillLutsky, @scottmac, @malthe, and @adamjstevenson, with angels from Anthropic, Stripe, Meta, Cursor, Replit, and more.
Frontier intelligence companies like @FireworksAI_HQ and @⬛⬛⬛⬛⬛⬛; fintech companies like @payabli and @vori; and healthtech companies like @thatch_ai use @falconer_ai because they all share the same vision as us: the knowledge you earn should belong to you.
Now you can know everything, so you can build anything.
Quant interns Summer '27
Optiver - 30 lpm
IITB - 16 hires
IITD - 13 hires
IITK - 12 hires
IITM - 11 hires
IITKGP - 4 hires
IITR - 1 hire
BITSP - 2 hires
MIT Manipal - 1 hire
IMC Trading - 25 LPM
IITB - 3 hires
IITD - 10 hires
IITM - 13 hires
IITK - 3 hires
IITKGP - 9 hires
IITR - 4 hires
Graviton - 25 LPM
IITB - 3 hires
IITD - 6 hires
IITK - 2 hires
IITKGP - 1 hire
Quadeye - 30 lpm
IITB - 5/6 hires
IITD - 5 hires
IITM - 1 hire
IITK - 7 hires
IITKGP - 5 hires
IITR - 2 hires
IITBHU - 1 hire
NK Securities - 25 LPM
IITD - 4 hires
IITK - 2 hires
IITKGP - 2 hires
IITR - 3 hires
IITG - 1 hire
IIT BHU - 1 hire
HRT - 56 LPM
IITB - 3 hire
IITD - 1 hire
IITKGP - 1 hire
IIITH - 1 hire
Jane Street - 40 LPM
IITB - 5/6 hires
IITD - 1/2 hires
Citadel Securities - 43.5 LPM
IITB - 4 hires
IITD - 2 hires
IITJ - 1 hire
Quantbox - 25 LPM
IITK - 1 hire
IITBHU - 1/2 hire
IITG - 1/2 hire (not that sure)
The above data might be off by 1-2 people but is mostly correct. Please feel free to correct in the comments.
Disclaimer - These are only for flagship infra/quant roles. Support roles like risk, swe are not included
If you’re a designer, here are better swaps than the obvious defaults:
Dribbble - Pinterest
Behance - Are na
Google Fonts - Fontshare
Unsplash - Lummi / Pexels
Coolors - Huemint
Figma Community - Layers / UI8
Mobbin - Refero
Awwwards - Godly
Notion templates - Super / Gridfiti
Canva - Figma
Adobe XD - Figma
Sketch - Figma
InVision - Framer
Marvel - Framer
Zeplin - Figma Dev Mode
Illustrator - Figma (for UI work)
Photoshop - Photopea
Lightroom - Darkroom
Blender - Spline
Cinema 4D - Spline
Webflow - Framer
WordPress - Webflow (for visual builds)
Wix - Framer
VS Code themes - Raycast extensions
Chrome bookmarks - Raindrop io
Stock icons - Phosphor Icons / Lucide
Stock illustrations - Storyset
LottieFiles - Framer Motion
Better task decomposition could be a major advantage for building reliable AI agents.
Aloha!🌺 Meet Ornith-1.5-122B: No Tradeoff Between Strong and Fast
Stronger. Faster. Leaner. Ornith-1.5-122B delivers frontier-level capabilities with exceptional inference efficiency. With superior instruction following, stronger task decomposition, and deeper understanding of real-world workflows, it achieves state-of-the-art performance among open-source models of comparable size, while running on a single NVIDIA DGX card!!
Open weights under the MIT license.
Coding
✅SWE-bench Verified: 0.845
✅TB2.1: 0.813
✅CyberGym: 0.800
✅NL2Repo: 0.512
Agentic
✅HLE: 0.446
✅BrowseComp: 0.785
✅MCP-Atlas: 0.746
✅Toolathlon: 0.708
🤗Hugging Face: huggingface.co/ornith-ai/Orn…
We can’t wait to see what you build with Ornith-1.5-122B.
Priya retweeted
imagine your AI agent hallucinating and spending $10k, who’s liable when it goes wrong?
exactly why companies are blocking them.
notice..
> Delta is blocking AI agents from booking flights
> JPMorgan Chase and Charles Schwab block agents from accessing accounts
> Yelp requires paid data licensing for agents
> Amazon has blocked Grok bot and other shopping agents.
the internet is starting to block AI agents that were supposed to do everything for us
Priya retweeted
MiniMax just dropped M3.1 Flash:
- 62.2% on automation-bench (beats opus 5 and sonnet 5.6)
- matches opus 5 and sonnet 5.6 on deepswe
- 200+ tokens/sec
- nearly 3x faster and 70% cheaper than m3
- costs $0.15 input / $0.60 output per million tokens (vs $5 / $25 for opus 5)
MiniMax M3.1 Flash: Space Performance + Bunny Speed
🚀🐰 Yes, Space Bunny - which has topped global leaderboards with 200T+ tokens - was an earlier version of M3.1 Flash.
💰Redefining the performance economics of agentic AI. M3.1 Flash achieves frontier-level intelligence across key tasks such as Cowork (Automation-Bench: 62%; ALE: 30%), Coding (DeepSWE: 74%) and Visual Reasoning (BabyVision: 91%), at 1/3 the cost of M3 and 3x the speed.
API pricing (flat rate): $0.15 / $0.60 per 1M input/output tokens, with cache reads at $0.015 / 1M tokens.
Model weights will arrive on Oct 23.
API: platform.minimax.io
M Plan: minimax.io/m-plan
MiniMax Code for desktop: agent.minimax.io/download
