Vikas Kulkarni R

It was very hard to previously exhaust @claudeai 20x max plan. This wasn't even the busiest week for me, I was on a vacation running basic automations and still hit the limits. Absolutely ridiculous limits!

@nieltenghu @gkisokay Even I want to understand Are there benefits to using Hermes over OpenClaw? Memory, is it that big of a deal because even OpenClaw has decent memory. is Hermes managing it better?

@gkisokay yeah, onboarding looks like the main friction with openclaw. Were you hosting hermes in the cloud or local ?

@nieltenghu I am non-technical, but its smooth. Talking with hermes feels productive since it has memory, so it doesn't feel like a chore. It feels like youre training a Pokémon.

Expectation vs Reality @zomato Yeh kya hai bhai 🥲 Accepting claude max subscription as a token of apology!

My YouTube : @vikaskulkarni.r?si=E8C2ggBK0dM_iuvl" target="_blank" rel="nofollow noopener">youtube.com/@vikaskulkarni…

I just watched my portfolio drop by ₹5 Lakhs in the final days of the financial year. I'd say this was one of the toughest year in the past 5 years of my trading career. 2023 was the previous toughest year for me (23% post charges and taxes). It's even lower this year with 21% post charges and pre taxes in trading. Investments are in soup with sizeable drawdowns. On the brighter side of things, what matters to me is my portfolio value at the end of the year and that went from 45 lakhs to 70 lakhs to now 65 lakhs because of the drawdown. But that's largely because of my income sources holding the portfolio steady. I'll be making a detailed video about how the FY went on my YouTube. Link in comments and bio!

Claude Max 20x - What are these limits? Switched back to 200k context window with better memory architecture. The quality degradation is significant after 500k and token utilisation is massive! What do you guys think?

@HNImanshu @claudeai @trq212 7 accounts! What's the plan on them?

@vikascoolkarnir @claudeai @trq212 I'm managing somehow with 7 accounts. Still I'm hitting rate limits every message. 😂

@trq212 @jasminewsun @TheAtlantic Finally you're online! What's up on the limits front, please tell me you've got some good news on the stability front 🥺

@jasminewsun @TheAtlantic congratulations! excited to read your work

Personal news: I’m joining @TheAtlantic as a contributing writer! It drives me nuts how wide of an understanding gap there is between SF AI world and everywhere else — especially given the immense public stakes. There's so much AI hype, anxiety, and misinformation; so doing translation and synthesis feels more important than ever. (This role is in addition to Subst*ck, where I’ll keep writing at the same cadence.) I'm using this excuse to share some rambly media thoughts: namely that tech journalism can & must be great again. The problem with “old media” is that it often refuses to take tech bros at their word, and the problem with “new media” is that it’s often just advertising, which is boring even for the subjects. There’s a doom loop where some reporters write poorly-informed stories, so insiders won’t talk to them, so sourcing is worse; not to mention that most journalists are not based in the communities they cover. This makes people bad-faith, but it also means a lot of AI reporting is 6-12 months behind. Yes, fantastic blogs/podcasts abound — these are the bulk of my info diet — but they are largely insiders talking to insiders, too niche to recommend to policymakers or smart non-AI friends. These fractures are a disaster for shared public knowledge, and make us less prepared to navigate AI well. Magazine writing offers the ability to rise above of the hourly play-by-play (squinting at every new model release, every new jobs report) and to the bigger questions. I actually think the most impactful AI writing has *months*, not days of longevity! Rather than over-anchoring to any particular forecast, it offers generalized frames for operating under uncertainty. A few types of pieces I’m especially keen to write: 1) AI culture: A few people’s idiosyncratic personal beliefs regularly change the world. It thus matters tremendously how AI builders view their work, politics, philosophy, and the future. I think most individuals in the AI industry are good and want their tech to do good. Journalists can portray AI workers’ earnest beliefs while being appropriately skeptical of how that can clash with or be shaped by industry incentives, and how it might diverge from the public. "Smart people confront hard moral/intellectual problem" is one of my favorite genres. 2) AI diffusion: AI discourse disproportionately focuses on its impact on software and writing because those are the jobs the messengers do (obviously I’m guilty of this). That makes me want to do more field reporting on AI in education, manufacturing, healthcare, etc: e.g. can I ride along with a team trying to integrate AI tutors into a school? Diffusion is rarely as smooth as economic models predict, and “how AI will go” depends largely on the speed, and where it hits first. Relatedly: AI in the non-western world. 3) AI superusers: Polls show people are highly anxious about AI’s speculative effects but sanguine about their personal use. I think more people should experiment with AI to feel both the pace of progress *and* its jagged edges. While AI can produce slop/surveillance/etc, it can also extend human ability & creativity. I want to paint portraits of people already “living in the future" so we can ask: is that a life we want? The tech is here, but we can choose how to relate to it. If you have ideas/feedback/etc my DMs are open, and my Signal is jws.27. For me 1-1 conversations are *not* on the record unless we say so. (I always thought this was a weird norm, and in general am happy to answer people's questions about “how journalism works” from my POV because it can be quite opaque.) (also I'm replacing my blurry macbook selfie with a b&w portrait profile picture to signify reluctant induction into the label of "capital-j Journalist.” I spent most of last year pretending to be funemployed, but I suppose this is graduation. end of an era!)

@claudeai lag gaye

I'm soo done with Claude. Too much cock for a bit of Personality, it's been quite a ride today. It's not working at all!

@trq212 claude setup-token is generating OAuth tokens (sk-ant-oat01-...) that return 401 "Invalid bearer token" immediately when used for API calls. Support told me to use API keys instead, but setup-token is the documented auth flow for self-hosted tools and it was working fine until recently. Is something broken on the OAuth side? Two accounts tested, same result. Please help me sort this out.

@Velazquez013 @calebhodges @openclaw Hey! Did you find a way to fix it?

@calebhodges @openclaw Hmm. I’ll run the update in the am, and try again.

Big news – the newest @openclaw update today quietly re-added oAuth for Anthropic Subscriptions! I just switched back to Sonnet 4.6 as my main without using the expensive API Token. Here's how in 60 seconds: Run "claude setup-token" in your terminal and save your token. Run "openclaw models auth add" Select anthropic → setup-token → paste Restart your gateway & open a new chat That's it. Your Pro/Max subscription now powers your local OpenClaw setup. No API key billing. Thanks @cathrynlavery for letting me know! @AndrewWarner We're back to Claude!

@kirubaakaran @iarjuntandon @OpenAI @openclaw Update it to 5.4 and see the personality!

@iarjuntandon @OpenAI @openclaw OpenClaw uses OpenAI Codex (gpt-5.2-codex) via OAuth, not the API directly. So am using the OpenAI model but through OpenClaw’s built in authentication, not by paying for API calls separately.

I have been using @OpenAI Codex with @openclaw, is Ollama more efficient ? Any views?

@trq212 It's amazing as long as you have unlimited access to Opus. The 1 million context window consumes tokens like water 🥲

i think we might have undersold 1M context tbh, the performance is so so good, I really just don't clear the context window much these days

@BandiShreyas The OG is back!

Ok, buying

@johann_sath Still worth it?

the openclaw file nobody talks about: HEARTBEAT.md this tells your bot what to do when you're not around. add this: "every 30 minutes: 1. check if any cron jobs failed 2. check if today's memory file exists 3. check disk usage 4. if anything broke, message me immediately" before: my bot sat there doing nothing while i slept. woke up to broken automations nobody told me about. after: bot runs 14 automated jobs overnight. if something breaks at 3am, i get a telegram message before i wake up. your bot should work while you sleep. not just when you're typing.

@steipete @openclaw I was coming to X and refreshing to only see this! Thank you

New @openclaw beta bits are up! Yes, includes GPT 5.4 and Gemini Flash 3.1! github.com/openclaw/openc…

Most AI skills are just brilliant communication

@trq212 @jxnlco @steipete This is beautiful! Wish you had something like that for students in India trying to build things using Claude... Rate limits are insaneee😔

@jxnlco @steipete 🫰 instructor actually goated

Got to work with @steipete and my main takeaway is I am not executing enough and I could sleep less.

The only thing going up today is India Vix