TimeCopilot

the agentic forecasting wave has arrived in Montreal at the International Symposium of Forecasting 🇨🇦💙 last year, we introduced TimeCopilot and our vision for agentic forecasting at the inaugural International Institute of Forecasters Open Source Forecasting Workshop and the first Women in Forecasting Conference. today, it’s a reality. a category of its own. organizations around the world are adopting agentic forecasting workflows, and a new frontier for time series forecasting is beginning to take shape. but new paradigms need rigorous evaluation. that’s why the TimeCopilot crew is excited to be in Montreal with two talks: 📈 Agentic Forecasting: Scaling Time Series Foundation Models in the Real World ⚡ Impermanent: A Live Benchmark for Temporal Generalization in Time Series Forecasting if you're in Montreal, reach out! Renée and i would love to grab coffee and talk forecasting, foundation models, benchmarks, the future of the ecosystem, and, of course, the important things in life. 🍃 and yes, we have a few more announcements planned for the week 👀 stay tuned. #ISFConf2026

we’re very pleased to welcome Datadog’s new Toto 2.0 family of models to TimeCopilot! 🔥🐶 this new family is a major step forward for time series foundation models, including one of the largest forecasting foundation models released to date with 2.5 billion parameters, strong performance in the GIFT-Eval benchmark by the Salesforce AI research team, and early signs of scaling laws: within the family, accuracy improvements appear to be positively correlated with model size. by adding Toto 2.0 to TimeCopilot, practitioners can now test whether those scaling laws hold for their own data and whether the models meet their accuracy and latency requirements. but not just that. TimeCopilot is the universal agentic framework for time series forecasting. practitioners can compare Toto 2.0 against other foundation models, deep learning and classical models, and simple baselines: in the same pipeline. without changing dependencies. with just one additional line of code. agenticly. ✨ and the best part? forecasting, cross-validation, anomaly detection, and prediction intervals are natively supported for the new Toto 2.0 family. and yeah: open source. 😎 let us know how it goes! tutorial in the comments. what would you like to see next? happy forecasting! 🫶

fly TimeCopilot, fly! 🦋 i’m tired. it’s been an intense journey. but i couldn’t be happier. here’s a quick recap of what the crew has achieved over the past few weeks: - welcomed Character Capital on board by joining Character Labs - crossed 25k downloads - became part of the top 25% most downloaded Python libraries - selected for PyCon US 2026 Startup Row in Long Beach, alongside 7 other startups shaping the future of software in Python - presented our paper at ICLR in Rio - working with amazing companies to integrate TimeCopilot into their products... more soon here  - published in International Institute of Forecasters' Foresight and featured by the forecasting community - building. building. building. and all of that while looking iconic. 💄 Renée and i couldn’t be more excited and grateful for everything that has happened, and especially for all the amazing people we’ve met along this early journey. and the stickers. we’re grateful to have stickers. and as forecasters, we’re already looking at the future. it looks agentic. :) more to come soon. fly TimeCopilot, fly! 💙

Startup Row at PyCon US 2026 is a wrap ~ and what a year it was! This program exists because the Python community believes in builders, and it is my honor to celebrate this year's cohort of eight companies taking eight bold bets on what Python can do next: Arcjet · CapiscIO · Chonkie · Pixeltable · SubImage (YC W25) · Tetrix · TimeCopilot · Phemeral Deep gratitude to the Python Software Foundation, our partners, the sponsors, my co-organizer Jason D. Rowley, and to every founder, volunteer, and attendee who makes PyCon the most human conference in tech. Here is to the ones building in public, shipping with purpose, giving back through open source, and doing it all in Python 🐍 #StartupRow #PyConUS #Python #Startups #OpenSource #AI #Innovation #Community https://lnkd.in/gbqqAWSN

what an amazing time Renée and i had at #PyConUS 2026 in Long Beach! 🏖️ TimeCopilot was selected as one of 8 startups for the Startup Row, and we had the opportunity to learn and share more about time series and Python with applied scientists, researchers, developers, founders, infrastructure engineers, and members of the open-source community. huge thanks to Jason D. Rowley and Shea Tate-Di Donna for organizing such a special event for the community 💙 this was my second PyCon and my second Startup Row, and i couldn’t feel more humbled seeing how far agentic forecasting has come in such a short amount of time, and to be building all of this together with Renée. and in the open. ✨ one thing became very clear during the conference, time series in the agentic era is still a nascent field, and there is an enormous opportunity ahead for the ecosystem. we can’t wait to share more of what we’ve been building at TimeCopilot. stay tuned! :) 🫶

New chapter for TimeCopilot. We've joined Character Labs, and we're so happy to have Character Capital on board. Super excited to work with Eli Blee-Goldman, Jake Knapp and John Zeratsky. The open, explainable forecasting ecosystem Azul Garza and I set out to build is starting to take shape faster than we expected. The momentum behind it keeps surprising us. We have more exciting updates to share very soon.

TimeCopilot is now one of the 25 most downloaded Python packages on the Python Package Index (PyPI) . 🎉 Good timing, with PyCon kicking off this week. The bigger signal is that the time series field continues to adopt TimeCopilot as one of its new tools. Azul Garza and I set out to build software that's measurable, reproducible, and open. Seeing practitioners and researchers reach for it confirms we're building the right thing. Thank you to everyone who's contributed, filed issues, and pushed us forward. We're just getting started. At PyCon this week? Come say hi.

TimeCopilot being one of eight startups selected for Startup Row at PyCon US still feels surreal. Since if you're like us, you probably love Python. If you're heading to PyCon US next week, come hang out with Azul Garza and me.

So excited to share that TimeCopilot just crossed 25k downloads. That's 20% growth from a couple of weeks ago. This year at ICLR in Rio, Azul Garza and I spoke with many time series professionals, and we can confirm that the landscape is shifting and evolving fast. We're thrilled that TimeCopilot is among the tools the community is embracing to gain speed, improve accuracy, and step into the agentic paradigm. As we keep growing, please keep sharing your ideas and thoughts on how we can make TimeCopilot better.