Shiva

Opinions are my own.

[Sai Rajeswar @RajeswarSai]

When we built EnterpriseOps-Gym, the goal was simple: stop measuring AI workers on toy tasks and start measuring them where enterprises actually live.

Today @ArtificialAnlys launches their implementation, EnterpriseOpsGym-AA, benchmarking frontier models across all 8 enterprise domains in a live, resettable environment. Independent, transparent, ecosystem-wide.

Grateful to this partnership. The gym is now everyone's.
https://t.co/Tb1Q2OaQpe
The article: https://t.co/AA8RgWXvWJ

@shiva_malay @siris_krishna @PShravannayak @TheJishnuNair @patricebechard @sagardavasam @ServiceNowRSRCH @srini994 @ysu_nlp @gneubig @chrisjpal @NVIDIAAI

[Sai Rajeswar @RajeswarSai]

📢 EnterpriseOps-Gym is now accepted to ICML 2026 🇰🇷✨

website: https://t.co/PJvBFnYsm9
🧩 1,150 expert-curated tasks
🏢 8 enterprise domains
🧰 512 tools
✅ Deterministic verifiers (Outcome + Integrity + Compliance)
📦 Fully containerized, no enterprise instance required

📊 𝗙𝗿𝗲𝘀𝗵 𝘂𝗽𝗱𝗮𝘁𝗲: GPT-5.5 numbers are out (alongside the strongest open and closed baselines). We will keep updating as new models drop because long-horizon reliability is moving fast, and we want to stay current.

One exciting (and humbling) signal: we’ve already seen frontier lab teams experimenting with EnterpriseOps-Gym to stress-test and improve their agents: including folks at OpenAI, Mistral AI, and NVIDIA AI. 🙏

📈 𝗘𝗮𝗿𝗹𝘆 𝗿𝗲𝘀𝘂𝗹𝘁𝘀 𝗮𝗿𝗲 𝗽𝗿𝗼𝗺𝗶𝘀𝗶𝗻𝗴: Top open models  including NVIDIA Nemotron Super  are showing strong performance, in some cases competing with frontier models.

@shiva_malay @sagardavasam @PShravannayak @turingcom @jonsid @ServiceNowRSRCH @Mila_Quebec

[James Rosen-Birch ⚖️🕊️ @provisionalidea]

it always amazes me when people don’t stop and think, “this person invented reinforcement learning. I’m just a rando. maybe *I’m* missing something.”

[🇺🇦 Dzmitry Bahdanau @DBahdanau]

🚨 New agent framework! 🚨

My team at @ServiceNowRSRCH is releasing TapeAgents: a holistic framework for agent development and optimization. At its core is the tape: a structured agent log.

Repo: https://t.co/IeQ8ufRUOh
Paper: https://t.co/BewNy7TkFA

Why you should care: 🧵

[Microsoft @Microsoft]

Bing.

[Libertarian Party @LPNational]

"I am not interested in picking up crumbs of compassion thrown from the table of someone who considers himself my master. I want the full menu of rights."

-#DesmondTutu

[Brad Smith @BradSmi]

Addressing the climate crisis requires broad collaboration and bold action.

[Brad Smith @BradSmi]

We’ve pledged as a company to become carbon negative by 2030. That’s why we’re moving towards a model where, every hour of every day, we’re matching our electricity consumption with zero-carbon energy purchases. https://t.co/EgpqXxTEoZ

[Microsoft Power BI @MSPowerBI]

Business outcomes should be driven by insights. Learn how enterprises can analyze their #data and use it to create actionable insights. Watch the webinar: https://t.co/2nL54HEuUd

[Libertarianism.org @libertarianism]

"Some will insist that animals are ineligible for political protection because, well, they’re not human. But what is it about being human that makes only humans worthy of protection?" https://t.co/hYrnWua5Af

[Microsoft Power Platform @MSPowerPlat]

Build apps fast with the tools built from the ground up for devs.

Get low-code tools that speak your language. Learn more: https://t.co/8ZRz3Yxycd #PowerApps #GitHub #Azure (1/2)

[Microsoft @Microsoft]

Not to brag, but @MicrosoftEdge blocked 95.5% of phishing attacks while Google Chrome blocked 86.9%.

Find out more about #MicrosoftEdge's highest-rated protection: https://t.co/05bpt2BpE1

[Global BC @GlobalBC]

BREAKING: A BC Liberal government will officially open up the auto insurance market to competition from private businesses and officially end the ICBC monopoly.

https://t.co/6p0gIghtEX

[Shiva @ShivaKrishnaRe3]

Wondering when Netflix is gonna make a documentary series on 2020 🤔

[zeynep tufekci @zeynep]

Next debate on zoom, with the "mute participant" option active.

[Microsoft @Microsoft]

We are the only cloud with limitless data and analytics capabilities that can deliver a cloud-native data estate for every organization. #MSIgnite

[Shiva @ShivaKrishnaRe3]

I would have broken up with @icbc years ago if I had a choice. But yet again, here I am being forced to tolerate another hundred million dollar loss. C’mon @SChandraHerbert @Dave_Eby @jjhorgan #BringChoiceBC #BCPoli

[Human Progress @HumanProgress]

Our 49th edition of #HeroesOfProgress spotlights Charles Babbage and Ada Lovelace, two 19th-century English mathematicians who pioneered early computing. https://t.co/ZQJnPOSD5g

[Human Progress @HumanProgress]

Few have heard the name Norman Borlaug, but his work saved the lives of one billion people and truly changed the fate of humanity for the better. #HeroesOfProgress

Full article: https://t.co/hZV6U3EcO6

[Cato Institute @CatoInstitute]

Forest tribes in India are not museum pieces to be preserved. To treat them as fundamentally different is paternalistic elitism. They deserve property rights and civil rights, and the opportunity to prosper. https://t.co/rGGo0xhzaY