KShivendu is in Chicago 👨🚀🦀
@KShivendu_
Followers
996
Following
5K
Media
129
Statuses
1K
building the future of search @qdrant_engine I love search, distributed systems, and llms · Creator of SmolDB
remote · blr
Joined March 2019
Launching my latest side project: smoldb 🐣 It's a distributed DB implemented from scratch in Rust. I'll keep on experimenting with this project to improve my knowledge of distributed systems, database, performance, etc ⚡
9
4
98
What's the best way to debug a deadlock/livelock? I take a gdb thread backtrace dump and cross reference it with the code to guess what might have caused it. But is there a better tool or approach?
0
0
1
Recently discovered that NYC has lots of cool startups. Is SF lagging in some sense?
0
0
3
We are sponsoring PyCon 🇮🇳 this year!
📷 We’re thrilled to announce our second Gold Sponsor – Qdrant We are happy to have @qdrant_engine contribute to the biggest #Python conference in the country. Read more in our blog: https://t.co/Jpt6bgiFGr
#sponsor #python #conference #bengaluru #gold
0
0
2
Now you can grep a PDF (and any document) Introducing SemTools - simple parsing and semantic search for the command line https://t.co/Zih7K4skqm
github.com
Semantic search and document parsing tools for the command line - run-llama/semtools
5
22
127
Look at the robots jumping together towards the end. This level of millisecond sync/coordination is extremely hard for humans. It's so good that it feels like VFX. Would be fun to watch Robots dance.
At Zhejiang University’s opening day, robot dogs are helping freshmen carry their luggage! 🦾🎒 They can haul up to 20kg, move fast, and even climb stairs with ease. https://t.co/7qZLybewSe
0
0
2
Rust folks try this to see magic: ➜ cargo clean Removed 50764 files, 60.8GiB total
pip install "huggingface_hub[cli]" huggingface-cli delete-cache Thank me later
0
0
2
Nice!
1
1
5
Please drop a 🌟 if you find it interesting :) https://t.co/Eto0DL2cIt
github.com
A smol distributed database implemented from scratch 🐣 - KShivendu/smoldb
1
0
4
OpenAI clearly hit a wall with LLMs. But it's kinda good. AI labs will have more incentive to understand the models to come up with new paradigms. And that should lead to a safer outcome for humanity.
0
0
2
https://t.co/ebaUXtFdDN Very interesting idea to solve alignment and improve models faster. Came into my radar after @drishanarora posted about @DeepCogito. We need the right framework to build AGI, data isn't the moat after a point.
ai-alignment.com
Guest post summarizing my approach to aligned RL.
0
0
1
Looks like Grok has been guessing all along. It's not based on real data.
@KShivendu_ Based on public interactions, as X keeps profile visits private, the most famous person likely to have visited is Sameer Nigam, founder and CEO of PhonePe. You tagged him in a 2022 post suggesting app improvements.
0
0
2
Hey, @grok, who was the most famous person to visit my profile? It doesn't need to be a mutual, don't tag them, just say who it was
2
0
1
Much awaited launch. Congratulations @anindyadeeps 🎉
The first AI-powered structure prediction editor, powered by Boltz-2 with bulk structure prediction, is now available on the LiteFold Platform. Links in comments. Exciting updates coming soon! 🚀
1
1
4
Brilliant execution by @sachdh!
Excited to share Aryabhatta 1.0, our leading model that scores 90.2% on JEE Mains, outperforming frontier models like o4 mini and Gemini Flash 2.5 Trained by us at @AthenaAgentRL , in collaboration with @physics__wallah, using custom RLVR training on 130K+ curated JEE problems
1
1
2
Agents aren’t reliable. They don’t learn from experience. At @composiohq, we provide skills that evolve with your agents @lightspeedvp gave us $25M to make agents usable
232
149
1K