Shyam V

Technology | Business | Engineering

[Shyam V @vorashm]

@Vtrivedy10 @mattwridley said “specialization” is what made all the human progress possible in his book the rational optimist. Looks like this is true for agents too.

[Shyam V @vorashm]

Was great seeing all the cool stuff @hwchase17 and team are building at @LangChain conference at Interrupt 2026!

[Harrison Chase @hwchase17]

"jev as a judge"

cheap and fast semantic verifiers! great for evals - especially online evals where you want to grade LOTS of traces

[Josh Elman @joshelman]

https://t.co/8zuACcwHGR

[Shreyas Doshi @shreyas]

Claude chat for those who seek a deeper understanding of this tweet:
https://t.co/boNuVo48TG

[Shyam V @vorashm]

looking to try @underdogdotai. who can invite me?

[Shyam V @vorashm]

@play2XP Still got an invite code?

[OpenAI Developers @OpenAIDevs]

Go from idea to a working agent faster with the Agents API.

Build and run cloud agents with the Codex harness, fully managed by OpenAI.

We handle orchestration, long-running sessions, and context management. You focus on what makes your agent unique.

Available in public beta.

[Shyam V @vorashm]

@MatthewBerman @bot @SpaceXAI Use grok bot to monitor your health vitals as well as grocery shopping, but making sure you’re ordering the right things to match your medical lab vitals. This is a huge market

[The Economist @TheEconomist]

Our analysis suggests that AI has so far created around 1m new jobs in America. We explain how the technology has created a hiring boom https://t.co/bywoXNJQ7A

[Shyam V @vorashm]

@MatthewBerman Moat*

[Shyam V @vorashm]

@MatthewBerman They no longer have most

[Shyam V @vorashm]

@MatthewBerman With your actual real email?

[Shyam V @vorashm]

@arena bug alert on Arena. Gemini 3.5 Flash high appears twice on the leaderboard

[Shyam V @vorashm]

@MatthewBerman Think it can beat fable in deepswe?

[Shyam V @vorashm]

@Kalshi Kalshi seems to be scamming people. Who are they to decide whether a mythos class model is not the same as mythos model? They’re playing with semantics here and making up their own rules. Also I don’t think they have any kind of customer support

[Shyam V @vorashm]

@Kalshi dafuq is this supposed to mean? This sounds like an outright SCAM

[Shyam V @vorashm]

The mobile app chat support doesn’t work either

[Shyam V @vorashm]

@Kalshi what is this anomaly about? The payout calc doesn’t make sense.

[Shyam V @vorashm]

@AnthropicAI ‘s latest model Opus 4.8 opens at rank 17 on @arena

So, it’s ranked below the following:

@Zai_org glm 5.1 (open weight)
@Google Gemini 3 flash
…14 others.

PS: it just debuted on the leaderboard recently and will likely go down further.