By
–
We just open-sourced FinQA — an #RL environment for financial reasoning agents. Real SEC 10-K data, multi-step reasoning + tool use, constrained SQL, binary rewards. The whole 9 yards! The kicker: a 4B model fine-tuned with FinQA outperformed a 235B model from the same family on finance reasoning: 58x smaller!
→ View original post on X — @snorkelai, 2026-03-31 00:19 UTC
By
–
Stop dumping 10 folders of context on your AI agent and wondering why it’s confused.
— This Week in Startups (@twistartups) 31 mars 2026
Instead, on board your agent like you would a new hire. Tell it who you are, what the job is, and point it to the right links.
Always be SPECIFIC about what you want, and let the agent figure… pic.twitter.com/iWekuort5M
Stop dumping 10 folders of context on your AI agent and wondering why it’s confused. Instead, on board your agent like you would a new hire. Tell it who you are, what the job is, and point it to the right links. Always be SPECIFIC about what you want, and let the agent figure out the rest for itself. More of Google AI PM @Saboo_Shubham_’s ultimate OpenClaw starter guide on today’s new TWiST… plus demos of @AgentMail and @MoltworldIO. Chapters : 0:00 Intro 1:24 Plaud: If your work depends on conversations — interviews, meetings, calls — you need a Plaud NotePin. You can check it out at Plaud.ai/twist and use code TWIST for 10% off! 3:05 We're Claw-pilled once again; it's an all AI Agent showcase 6:15 Google AI PM Shubham Saboo's Top 5 OpenClaw tips 10:14 Quo (formerly OpenPhone) gives you a clean, modern way to handle every customer call, text, and thread all in one place. Try it free at quo.com/TWiST. 13:26 Tip #1 — Onboard your agent like a new hire 17:31 Tip #2 — Talk to your agents constantly 19:02 Tip #3 — Put your agents on a schedule 19:51 LinkedIn Jobs – Hire right, the first time. Post your first job and get $100 off towards your job post at LinkedIn.com/twist. 23:41 Tip #4 — Add cross-agent memory 29:33 Tip #5 — Let your agents self-improve 29:51 Iru unifies identity, endpoint security, and compliance into one platform. TWiST listeners get 20% off when they book a demo at iru.com/twist! 34:03 Why Jason says founders should avoid journalists (cc: @Jason, @Lons, @Haakamaujla) 🎥 Watch the full episode here 👇
→ View original post on X — @saboo_shubham_, 2026-03-31 00:11 UTC
By
–
The first @TheOfficialACM conference on agentic AI systems just got a boost. @SnorkelAI is joining as a sponsor of @CAISconf this May in San Jose. Stanford AI Lab roots, production AI focus, and a shared belief that this community needs a rigorous home. caisconf.org
→ View original post on X — @snorkelai, 2026-03-31 00:09 UTC
By
–
Must-read AI research of the week: ▪️ Learning to Commit: Generating Organic Pull Requests via Online Repository Memory ▪️ Effective Strategies for Asynchronous Software Engineering Agents ▪️ Composer 2 ▪️ From Static Templates to Dynamic Runtime Graphs: A Survey of Workflow Optimization for LLM Agents ▪️ Scalable Prompt Routing via Fine-Grained Latent Task Discovery ▪️ MSFT: Addressing Dataset Mixtures Overfitting Heterogeneously in Multi-task SFT ▪️ On the Direction of RLVR Updates for LLM Reasoning: Identification and Exploitation ▪️ Sparse but Critical: A Token-Level Analysis of Distributional Shifts in RLVR Fine-Tuning of LLMs ▪️ Why Does Self-Distillation (Sometimes) Degrade the Reasoning Capability of LLMs? ▪️ RL for Distributional Reasoning in LMs ▪️ Rethinking Token-Level Policy Optimization for Multimodal Chain-of-Thought ▪️ EVA: Efficient Reinforcement Learning for End-to-End Agent Find the full list and the main AI news here: turingpost.com/p/fod146
→ View original post on X — @debashis_dutta, 2026-03-30 23:41 UTC
By
–
In the FinQA env, a 4B model was fine-tuned to outperform a 235B model from the same family on our Finance Reasoning benchmark. What did we teach the 4B model? Tool discipline. Learn more: snorkel.ai/blog/building-fin…
→ View original post on X — @snorkelai, 2026-03-30 22:51 UTC
By
–
Our FinQA environment is available on OpenEnv (s/o @huggingface + @PyTorch) FinQA is an open RL environment with: • 290 expert-curated questions • Real SEC 10-K data • Tasks requiring multi-step tool use RL proof point on FinQA: make a 4B model > 235B model 👇
→ View original post on X — @snorkelai, 2026-03-30 22:51 UTC
By
–
AI: Here's your 90-day roadmap to execute on!
— The Rundown AI (@TheRundownAI) 30 mars 2026
User: Build it literally right now.
AI: pic.twitter.com/wxkE55lil8
AI: Here's your 90-day roadmap to execute on! User: Build it literally right now. AI:
By
–
This is an insane Anthropic tweet. And it’s a *buried reply* to one of their other tweets. I am reminded of a talk I gave ~2-3 months ago where a senior developer at a Fortune 500 company asked me “why would I use AI to code if I can just code myself.” I answered. He said, “But sometimes it messes up.” I told him this was coming. Even if it’s not perfect today (it makes weird product features decisions sometimes, not gonna lie), the scaling laws seem to be holding up this year and the next iteration will be even more capable. I wish I could send him this tweet.
→ View original post on X — @alliekmiller, 2026-03-30 21:26 UTC
By
–
We’re trying to build intelligent systems… using control frameworks designed by humans. That’s the core limitation of today’s agent harnesses. A new paper from Tsinghua University and Shenzhen proposes something radically different: 👉 What if the harness itself is not code—but natural language? Instead of hardcoding orchestration logic, they introduce Natural-Language Agent Harnesses (NLAH): – The control logic is written as an editable natural language SOP – The LLM interprets and executes that SOP dynamically – A shared runtime enforces structure via contracts, artifacts, and adapters Even more interesting: ➡️ The SOP itself can be generated and adapted by AI depending on the task So instead of: > Humans define → Agents execute We get: > AI defines → AI executes → AI evolves 🧠 Technical takeaway This shifts agent design from: – Static orchestration graphs – Hardcoded tool pipelines – Rigid planner-executor loops To: – Executable natural language control logic – Runtime-interpreted orchestration – Portable, composable harness artifacts The harness is no longer buried in code—it becomes a first-class abstraction. 🏗️ Architecture implications – Decouple control logic from implementation – Treat orchestration as data, not code – Use LLMs as meta-execution engines – Design systems that scale with tokens, not constraints 💡 Bigger question If agents can define and execute their own control logic… What else in AI system design should stop being code—and start being language? 📄 Paper: arxiv.org/abs/2603.25723 🔗 Follow my communities and personal initiatives: • Amazing AI, Data, Quantum Computing & Emerging Technologies — drdebashisdutta.com/ • Research & Innovation – Quantum, AI & Advanced Systems — researchedge.org
→ View original post on X — @debashis_dutta, 2026-03-30 21:25 UTC
By
–
🦞👀👏🏾 Could OpenClaw be the next ChatGPT?
— Catherine Adenle (@CatherineAdenle) 30 mars 2026
In an interview with Jim Cramer, Jensen Huang says OpenClaw could unlock a whole new wave of AI.
➡️ Not just chat.
➡️ AI that acts.
We’ve moved from:
Search ➝ Generate ➝ Execute
It’s a big shift and I am here for it.
💬 What’s your… pic.twitter.com/ZnCyG6BD7t
🦞👀👏🏾 Could OpenClaw be the next ChatGPT? In an interview with Jim Cramer, Jensen Huang says OpenClaw could unlock a whole new wave of AI. ➡️ Not just chat. ➡️ AI that acts. We’ve moved from: Search ➝ Generate ➝ Execute It’s a big shift and I am here for it. 💬 What’s your take, real breakthrough or overhyped? 📊 source: madmoneyoncbc | IG 📌 Tags: #AI #NVIDIA #ChatGPT #Tech #OpenClaw #TechNews #Innovation #GenAI #FutureOfWork #ArtificialIntelligence #Technology #MachineLearning #Startups
→ View original post on X — @catherineadenle, 2026-03-30 21:19 UTC