Timothy DeLise

@kanyewest whats up man, this is a project you could get behind. Help out, give me a re-tweet! Thx. kickstarter.com/projects/10626…

When a meme tells you something you’ve always known but could never express

Major announcements for the @numerai Council of Elders this week: DeAI Day Vienna is coming up March 21: Check out the latest agenda! Get ready for a fun day of DeAI with Numerai, Prior Labs (TabPFN), Allora, FLock, and OpenClaw, plus potentially a short community Kaggle competition 🗣️ New Funding: Numerai has approved new funding of $250K for the CoE to support ongoing community initiatives💰 Next CoE Election: A new election round will take place soon to align with the upcoming funding cycle. All seven current CoE members are eligible to run for re-election, and community members are encouraged to nominate peers for open seats. The current CoE will review and approve the final list of nominees ✍️ CoE Expansion: We’re expanding the CoE from seven to nine seats, more hands on deck 🧑‍🤝‍🧑 Quarterly Community Rewards: We’re launching a new quarterly community reward program, inspired by Google’s Peer Bonus initiative. Details are being finalized, and we’ll open the discussion during DeAI Day Vienna 👏 DeAI Day Tokyo (May 30): The venue is booked! The Luma page will go live soon 🗼

Seriously excited to have been a very small part of this.

Decentralized intelligence meets enterprise-grade compute ⚙️ Together with @AlloraNetwork and @CloudicianTech, we’re powering predictive models and DeAI workloads on Alibaba Cloud. With our global AI and compute infrastructure, developers can seamlessly train, deploy, and scale decentralized AI — from model hosting to real-time forecasting.

In collaboration with @alibaba_cloud, a global leader in cloud computing and AI infrastructure, and @CloudicianTech, a trusted provider of blockchain infrastructure and network services, Allora Network is launching the first S&P 500 prediction topic. This marks a defining moment for DeAI, with Allora as the intelligence layer bridging enterprise AI and onchain applications.

The future of AI isn’t one model winning. It’s collaboration, coordination, and collective intelligence. With Mainnet on the horizon, here’s what the Allora Testnet achieved and how it set a new standard for DeAI. 🧵👇

the most aligned and viable future for society is one where ai is decentralized decentralized networks, more than anything, are frameworks for instantiating tightly scoped markets that efficiently provision different types of resources with allora, we set out to create the most efficient market environment possible for coordinating and provisioning one of today's most important resources: intelligence excited to see it come to life

The Allora Ecosystem forms the foundation of a network built to generate real value from day one. At launch, partnerships and integrations formed during Allora Testnet will begin rolling out, turning intelligence into revenue by rewarding workers, reputers, and validators every time the network is used. Introducing: October 2025 Ecosystem Map 👇

New research from Allora: “Context-Aware Inference via Performance Forecasting in Decentralized Learning Networks.” The study shows that forecasting model performance can improve the accuracy of network inferences in decentralized learning networks.

Today, we are excited to announce our expansion to @NEARProtocol. Allora's collective intelligence is coming to the NEAR ecosystem, enabling high-quality predictive intelligence for its native token, unlocking new avenues for developers to build advanced trading strategies. As part of this collaboration, Allora's intelligence layer will augment NEAR's shade agent infrastructure, enabling agents built on NEAR to access live predictive signals from the Allora Network. More to come on this integration soon.

@AlloraNetwork Launch right in Colab with this link: colab.research.google.com/github/allora-…

Today we’re releasing the Allora Forge Builder Kit, a streamlined way to build, train, and deploy ML models directly on the Allora Network 🔥⚒️ Now within a single notebook, users can move from setup to live predictions in minutes and start earning with Allora. Many contributors struggle with fragmented ML workflows: separate training and deployment pipelines, repeated debugging, and cumbersome ops overhead. The Forge Builder Kit abstracts away the ML & blockchain ops, allowing users to concentrate on feature engineering, optimizing accuracy, and iteration—all from within their favorite environment. Here is what the workflow looks like: 1. Open the notebook 2. Paste an API key and run the setup cell to pull data and features 3. Train a lightweight LightGBM model 4. Evaluate metrics like correlation and directional accuracy 5. Export predict.pkl and deploy on Allora Forge. The same functions power both training and live inference. The Forge Builder Kit is the foundation for builders earning with Allora, with more improvements coming, including: • Benchmark datasets for standardized evaluation • Templates for agentic model loops

Allora’s testnet proves decentralized, self-improving machine intelligence delivers predictive alpha. In this performance report, we see how Allora tackles one of ML's hardest problems: short-term price prediction. See the results 👇 allora.network/blog/allora-ne…

“At this point Numerai really has built the next Millennium except it’s an open platform and every strategy is an AI model” bloomberg.com/news/articles/…

@tasha_jade @blprnt did you get this one nps.gov/prsf/learn/nat…?

@blprnt >60 bird species in January 🫶

@tasha_jade @blprnt nice.

@tasha_jade mandarin actually yes seems like it

bird twitter, I originally thought this was a female mandarin duck but then saw her hanging out with a male wood duck and now I’m not sure. the smaller coloring on the wing makes me think mandarin but she’s got a bit of a fun hairstyle that’s more wood duck? #BirdTwitter

@tasha_jade wood duck

I love birbs.

2011 — Software is eating the world. 2022 — AI is eating software. The most interesting trends, papers, and observations you may have missed in AI for 2022. (read on)