ARiss

@ankurnagpal I have a working application. Let’s set up a demo

A big new consumer category will be an AI house manager for the entire family Can live in a group chat and can handle tasks like: - Manage groceries - Plug into home security - Scheduling maintenance & cleaning - Smart home controls and automations Who's building this?

You know how some people seem to have a magic touch with LLMs? They get incredible, nuanced results while everyone else gets generic junk. The common wisdom is that this is a technical skill. A list of secret hacks, keywords, and formulas you have to learn. But a new paper suggests this isn't the main thing. The skill that makes you great at working with AI isn't technical. It's social. Researchers (Riedl & Weidmann) analyzed how 600+ people solved problems alone vs. with an AI. They used a statistical method to isolate two different things for each person: Their 'solo problem-solving ability' Their 'AI collaboration ability' Here's the reveal: The two skills are NOT the same. Being a genius who can solve problems in your own head is a totally different, measurable skill from being great at solving problems with an AI partner. Plot twist: The two abilities are barely correlated. So what IS this 'collaboration ability'? It's strongly predicted by a person's Theory of Mind (ToM)—your capacity to intuitively model another agent's beliefs, goals, and perspective. To anticipate what they know, what they don't, and what they need. In practice, this looks like: Anticipating the AI's potential confusion Providing helpful context it's missing Clarifying your own goals ("Explain this like I'm 15") Treating the AI like a (somewhat weird, alien) partner, not a vending machine. This is where it gets strange. A user's ToM score predicted their success when working WITH the AI... ...but had ZERO correlation with their success when working ALONE. It's a pure collaborative skill. It goes deeper. This isn't just a static trait. The researchers found that even moment-to-moment fluctuations in a user's ToM—like when they put more effort into perspective-taking on one specific prompt—led to higher-quality AI responses for that turn. This changes everything about how we should approach getting better at using AI. Stop memorizing prompt "hacks." Start practicing cognitive empathy for a non-human mind. Try this experiment. Next time you get a bad AI response, don't just rephrase the command. Stop and ask: "What false assumption is the AI making right now?" "What critical context am I taking for granted that it doesn't have?" Your job is to be the bridge. This also means we're probably benchmarking AI all wrong. The race for the highest score on a static test (MMLU, etc.) is optimizing for the wrong thing. It's like judging a point guard only on their free-throw percentage. The real test of an AI's value isn't its solo intelligence. It's its collaborative uplift. How much smarter does it make the human-AI team? That's the number that matters. This paper gives us a way to finally measure it. I'm still processing the implications. The whole thing is a masterclass in thinking clearly about what we're actually doing when we talk to these models. Paper: "Quantifying Human-AI Synergy" by Christoph Riedl & Ben Weidmann, 2025.

#AI right now won’t replace you. But the person who uses AI will. In 10 years, ignoring it won’t just be a mistake—it’ll be extinction. #ai #ML #genai

@AngelHack @MadeinSWPA Tiny details - big impact - you’re welcome to check it out

@AronRiss @MadeinSWPA The butterfly and flower designs are stunning! Such intricate nail art.

Thanks to @MadeinSWPA for co-hosting! Cut #downtime; fix it right the first time 🏭 See how Korra.ai’s visual-first AI turns manuals, drawings & SOPs into source-cited answers on the line. link: catalystconnection.org/event/catalyst… #manufacturing #ai #reliability #secureAI

@Shpigford @Namecheap Heard of @GoDaddy ?

awful @namecheap support experience on a premium domain purchase. taking 100+ domains elsewhere. where should i transfer all of them to?

@Shpigford WOM starts in-product. When someone hits the “aha,” hand them a brag pack to share in 10s: 📈 win chart Δmetric, a 1-liner, and a pre-filled Slack/Email to a peer, plus a referral link that gives them credit. Helpful for them; trackable for you.

How do you encourage more word-of-mouth from a marketing perspective? (Context is B2B SaaS)

@omarsar0 Impressive work—seeing a 14B agent with agentic RL push math performance is a real data-efficiency story. The code-tool loop seems key. Curious: how much lift is RL vs tool-use vs prompt curriculum, and does it transfer beyond math?

Quick Overview rStar2-Agent (Microsoft Research). A 14B math-reasoning model trained with agentic RL that learns to think smarter by using a Python tool environment, not just longer CoT. It introduces GRPO-RoC, a rollout strategy that filters noisy successful traces, plus infrastructure for massive, low-latency tool execution.

Cool research from Microsoft! They release rStar2-Agent, a 14B math reasoning models trained with agentic RL. It reaches frontier-level math reasoning in just 510 RL training steps. Here are my notes:

@connordavis_ai @Scobleizer Hybrid feels right: SLMs for repeatable skills; LLM only for novel pivots. Add a self-check line + uncertainty per step to stop error cascades (cheap auditability)

The most important AI paper of 2025 might have just dropped. NVIDIA lays out a framework for Small Language Model agents that could outcompete LLMs. Here’s the full breakdown (and why it matters):

AI transformation isn’t a switch—it’s a journey. 🚀 The best time to start was yesterday. The second-best? Right now. #ai #transformation #consulting #genai #promptengineering #vibecoding #llm #aicommunity #gang

👋 I #build #AI that drives business. Early team @ThetaRay (AML/cross-border), then IronVest, Helport.ai, Korra.ai. Now I partner with founders & execs to ship prompts, playbooks, and proofs. If you can dream it, you can prompt it #prompt #Ai

New here—I build lightweight AI agents for B2B SaaS. I post: agent design, RAG that doesn’t break, evals, and shipping checklists. If you’re an engineer, founder, or PM, follow for practical systems and templates. #aiagents #b2bsaas #growth #ai

@stockx Any update? It’s been 21 days....

@AronRiss Hey, thank you for providing this information. Our team will follow up via email as soon as possible. We appreciate your patience.

@stockx any update on my 900 dollar order that has taken over 20 days???? 16369507-16269266. That’s my order number make it happen or I’m switching to @eBay.

@stockx are you still in business? My package is 16369507-16269266. Is long overdue. Get back to me ASAP??? This is absolutely unacceptable. Get your packages together.

@stockx I spent 900 dollars and it’s been over 20 days! And still no package???? Does @stockx even work? Should I have used @eBay?

@Xbox help! I have the Xbox series c in my cart but can not purchase it! Help!@Microsoft

@optimum @OptimumHelp No one answers your phone, your tweets. Is any one still with the company? I have been trying to get ahold of any one from your organization for 2 weeks. I need a new MODEM ASAP

@OptimumHelp 2 weeks of trying to get ahold of you!!! I have had two reps come to my house, and they said they cant order a modem!! please have someone contact me. this is unacceptable.

@sellis13 I HAVE BEEN TRYING TO GET AHOLD OF @OPTIMUM FOR TWO WEEKS. NOT SURE ANY ONE WORKS THERE??

I have doesn’t 2 days trying to get through to customer service at Optimum Altice 1 st day I asked for a call back that did not come 2nd day I stayed on hold for 1 hour 55 minutes they picked up and hung up ...nice service Hey Optimum you can call me #optimum #Cablevision