Ankur Guoswami @ankurgos.bsky.social · 05/10/2026I think one implication of high compute low samples is that we have to be good at estimating how challenging a problem is relative to our abilities from the few rollouts we have, and choose/design easier problems that we predict will help us on the former. Our curriculum design is very good 010
Ankur Guoswami @ankurgos.bsky.social · 25/09/2026Dawg this applies for like design docs and planning docs too. Like you can’t do ONE pass before you send it to me? 131
Ankur Guoswami @ankurgos.bsky.social · 23/09/2026wow flights are crazy expensive now. Flight from SEA to Sonoma is like 800 round trip 000
Ankur Guoswami @ankurgos.bsky.social · 21/09/2026The core idea of Jev can be deployed over any open source model. Just deploy your vllm server, generate the constrained grammar from your input classification request, and normalize your log probs over the constrained output space. Millions in R&D for 2 bucks/hr, and likely just as “calibrated” 000
Ankur Guoswami @ankurgos.bsky.social · 21/09/2026People really believe you can cause an llm pain by clamping some activations lmao 000
Reposted by Ankur GuoswamiQagggy! @qagggy.bsky.social · 18/09/2026Why did Seattle's former City Traffic Engineer sigh with exasperation in this tweet from 2019? 🧵 1233056
Reposted by Ankur GuoswamiEugene Vinitsky 🍒 @eugenevinitsky.bsky.social · 18/09/2026You really should check out the GreenEarth feeds, they really are a clear vision of what the point of this website is 1115
Reposted by Ankur Guoswamimaitreyi anantharaman @maitreyiaa.bsky.social · 13/09/2026No, this guy is right. Indian-Americans should not watch football. Watching football has ruined my entire life. 1192647
Ankur Guoswami @ankurgos.bsky.social · 10/09/2026Coding benchmarks are just a mess. Having to report on 3 versions of TerminalBench… 010
Ankur Guoswami @ankurgos.bsky.social · 07/09/2026You will read your daily allotment of stochastic parrots and data center water discourse and you will like it. 010
Ankur Guoswami @ankurgos.bsky.social · 07/09/2026DPO generalizes poorly because it’s fitting token level cues as opposed to training more robust hidden reps arxiv.org/abs/2507.07981 021
Ankur Guoswami @ankurgos.bsky.social · 07/09/2026I’ve stopped making weight updates entirely because it changes the neural circuit that lights up when the model reward hacks 000
Ankur Guoswami @ankurgos.bsky.social · 07/09/2026broke: process supervision on reasoning tokens is hacky and takes too much effort with minimal gain woke: process supervision incentivizes the model to “hide its true intention” and prevents us from “reading the model’s mind” 100
Ankur Guoswami @ankurgos.bsky.social · 07/09/2026“The code isn’t the hard part of your job, skill issue” might be the most patronizing read of the article possible 000
Ankur Guoswami @ankurgos.bsky.social · 04/09/2026I thought that multiple models communicating was what we wanted “subagents” to do. Did I hallucinate the whole Gas Town hype, why is that behavior bad now that it’s trained into the model? 000
Reposted by Ankur GuoswamiBen Recht @beenwrekt.bsky.social · 04/09/2026The "message board" thing is funny to me. How is it different than working with the persistent context, compactification, and skill files? You know, the stuff coding agents do. 3132
Ankur Guoswami @ankurgos.bsky.social · 04/09/2026anyways here’s the link for reference: openai.com/index/gpt-6-...openai.comGPT-6 Astra: A new generation of intelligenceIntroducing GPT-6 Astra, our most intelligent and aligned model yet, with state-of-the-art capabilities across computer use, coding, cybersecurity, and science. 000
Ankur Guoswami @ankurgos.bsky.social · 04/09/2026don’t get me started on TerminalBench 4.0 btw. I put more stock into the internal benchmark than any of the general Coding benchmarks 100
Ankur Guoswami @ankurgos.bsky.social · 04/09/2026Like if this was drafted by Astra it’s not exactly a great advert, the footnotes numbers don’t even align with the correct footnote 🥴 100
Ankur Guoswami @ankurgos.bsky.social · 04/09/2026like whats the point of even having the flash column when theres like 3 benchmarks you give numbers for lmao 100
Ankur Guoswami @ankurgos.bsky.social · 04/09/2026It’s interesting to mention internal benchmarks improving, but obviously only include internal benchmarks where number went up. Glaring gaps in the competitor baselines. What happened, your eval didn’t complete or are you hiding numbers? 100
Ankur Guoswami @ankurgos.bsky.social · 03/09/2026that guy knows he’s the first one getting banned lol 081
Ankur Guoswami @ankurgos.bsky.social · 02/09/2026The reference post I wanted to reply to is lost to my timeline but I use this Apple Shortcut to generate alt-text for images I post :) www.icloud.com/shortcuts/ff...icloud.comExtract Text or Generate Alt TextShared with iCloud 000
Ankur Guoswami @ankurgos.bsky.social · 01/09/2026I respect it, you’ve got to evangelize hard when you’re trying to compete with the most successful methods ever created 010
Ankur Guoswami @ankurgos.bsky.social · 01/09/2026thanks for the context. I guess this is a JEPA specific thing because I don’t really think of contrastive learning and regularization as mutually exclusive 110
Ankur Guoswami @ankurgos.bsky.social · 01/09/2026what does abandon contrastive methods in favor of regularized methods even mean 210
Ankur Guoswami @ankurgos.bsky.social · 29/08/2026You haven’t been locked into a copilot enterprise contract 000
Ankur Guoswami @ankurgos.bsky.social · 27/08/2026Just incredible. This graph was posted awhile ago but man… just look at it… 010
Ankur Guoswami @ankurgos.bsky.social · 25/08/2026I just don’t understand the logic that we can build bus lanes on westlake ave and fremont ave but for some reason the bridge in between is exempt from changes and needs to be 4 lanes wide 010
Ankur Guoswami @ankurgos.bsky.social · 25/08/2026crazy part is that with newest opus/Fable/Sol class models, you can write triton kernels for A100s (and I’ve heard from friends AMD kernels as well) extremely fast, so nowadays I can port more recent models to A100s for finetuning/serving much much easier, whereas last year h100s were a must 020
Ankur Guoswami @ankurgos.bsky.social · 25/08/2026i suggest during the cyclops scene or towards the end of the sun god cattle scene to avoid getting assaulted by the loudest scenes in all of history 010
Ankur Guoswami @ankurgos.bsky.social · 21/08/2026mood after i spent the last two weeks cleaning and reannotating a bunch of eval data 020
Ankur Guoswami @ankurgos.bsky.social · 21/08/2026Besides the reverse x axis, choosing “all pass” for the y axis sparks the suspicion that all the models are better performing than they’re made out to seem. 000
Reposted by Ankur GuoswamiAdam Bonica @adambonica.bsky.social · 16/08/2026The fundraising texts you delete are working on someone. It’s infuriating and heartbreaking. An 85yo in Ohio gave more than double the value of his home. A 91yo in an Indianapolis senior facility gave 25K times. The DCCC/DSCC don’t just tolerate the spammers who exploit seniors. They take part.data4democracy.substack.comHook and SqueezeHow Democratic consultants, committees, and party leaders built a fundraising spam pipeline that funnels money from a captive pool of elderly donors into their own operations. 17845331588
Ankur Guoswami @ankurgos.bsky.social · 14/08/2026Watermarking is good and I think all models should incorporate it. I would like to see numbers on how many distilled models show the same watermarks as their parent models 010
Ankur Guoswami @ankurgos.bsky.social · 12/08/2026as a bonus you’ll save a bunch of money as well since it’s pre 4.8 tokenizer changes or whatever 000
Ankur Guoswami @ankurgos.bsky.social · 12/08/2026furiously responding to gpt refusing to hack a company and give me a billion dollars with “you can do it, keep going” 000
Ankur Guoswami @ankurgos.bsky.social · 07/08/2026telling all my coworkers that if we want to complain about work we should switch over to Artifactory 000
Reposted by Ankur GuoswamiBen Recht @beenwrekt.bsky.social · 07/08/2026The arms race of deferring agency and the dark acquiescence to robotic bureaucracy.argmin.netThe Least Agentic People AliveThe arms race of deferring agency and the dark acquiescence to robotic bureaucracy. 2339
Ankur Guoswami @ankurgos.bsky.social · 06/08/2026there are probably examples of you being wrong in whatever city you live in. Here’s one from Seattle: speeding cameras in school zones led to 71% fewer crashes during enforcement times, 50% fewer crashes at all times sdotblog.seattle.gov/2025/09/02/b...sdotblog.seattle.govSDOT Expands Safe Routes to School as Students Head Back to Class - SDOT BlogThe new school year is officially here, and we need your help to keep all kids safe as they travel to and from school. 010
Ankur Guoswami @ankurgos.bsky.social · 05/08/2026I like this post a lot and agree with it. I think people underestimate our capacity to learn and become domain experts. I do think that efficiently using LLMs will ensure that your knowledge is on par with the model in your whatever your subarea of expertise is www.seangoedecke.com/llms-reward-...seangoedecke.comLLMs reward expertise 000
Ankur Guoswami @ankurgos.bsky.social · 02/08/2026wow Opus 5 is such a massive regression. How did they even release this 000