Sagar
293 posts
Sagar
@sagarck04
ml @columbia, prev @join_ef, building!
new york Katılım Kasım 2022
127 Takip Edilen129 Takipçiler
Sagar retweetledi
Ara is the self-driving IDE.
Instead of waiting for your next prompt, it ships features for you. @aradotso maintains self-improving memory across the app, Claude, and Codex.
Congrats on the launch, @xadisingh & @svemyhre!
ycombinator.com/launches/Qcm-a…
English
@gill_kyle one time i saw a business generate their logo, it said "freshly squeezed lemons" three separate times.
English
Spent a bunch of time redesigning our filtering UI. Natural language first. Highlights and hover states to show query understanding
English
not sure from where kids today are getting these ideas.
I'm sure I'll have much better things to talk about.
English
i fixed luma search.
i tried finding my friend’s hackathon in nyc and literally couldn’t.
it wasn’t tagged tech or ai, so luma basically hid it. he had to send me the link directly.
made this to search events in nyc + sf.
syncs every hour:
fix-luma-search.vercel.app
English
first impressions of grok build by xai and i already love the vibe. clean tui that's fast and interactive clickable, the token allocation diamond visual is genuinely a nice touch.
signed in with x premium+ and it just worked out of the box. 512k context window, 28 tools, auto compact at 85%. ill keep testing and reporting back.
Sudo su@sudoingX
anyone at xai on the grok build team, ping me for my peace of head, peacefully testing the model and grok keeps confidently telling me this same thing every session. that someone on your side provisioned my account with internal/partner tier (tier 4 + team + grok-cli scope) tied to nousresearch contributor status + the active grok provider pr, parallel internal distribution path that lets normal x login unlock the full agent binary without going through the supergrok heavy route. either xai actually did this and grok is leaking the lore, or grok is hallucinating a very specific story about me. both are interesting, just tell me which. ill keep building either way.
English
cursor cli is so fucking fast it's unreal. if you jumped from claude code the difference is not even close. the speed alone changes how you think about prompting. i'm shipping faster than i ever have.
Sudo su@sudoingX
cursor cli with composer is the killer combo rn.
English
We should start training model directly from neural signal. Human will one day talk to machine through brain wave. The communication bandwidth of typing/speaking is the bottleneck.
Elon Musk@elonmusk
@XFreeze Neuralink is a much bigger breakthrough than most people realize. Enabling people to control a computer with their mind and the completely blind to see are Jesus-level miracles.
English
English
Some enterprise tasks are challenging to hill-climb with RL-based methods since they involve very out-of-distribution behavior. On-policy self-distillation (OPSD) gives a model learning signal for every token it writes, far richer than the single scalar reward of RL.
But that channel is noisy: most tokens don't reflect the behavior you're after. We introduce Relevance-Masked Self-Distillation (RMSD), which uses a two-step filtered loss mask to cut through the noise and find the tokens with the highest signal. Compared to OPSD it trains more stably, provides higher data efficiency, and reaches a higher performance ceiling.
English
I created a list of my favorite cafes to work at in NYC.
Coffee shops > offices.
I started my company inside a coffee shop.
Beautiful spaces. Good energy. Surrounded by people locking in. And you randomly meet the most interesting people.
My list includes the cafe name, neighborhood, and a proprietary, confidential scoring system based on:
- Work space (tables, outlets, WiFi, design)
- Food (quality, options, price)
- Music (playlist, can you take calls)
- People (do interesting people go here)
- Coffee (does it hit)
- Vibes (overall energy)
I'd love to share this list with you + add new spots.
Comment "CAFE" and I'll DM you the list.
English