An empty shelf looks exactly like weak demand to a forecasting model. You sold nothing because you had nothing, and it tells you to order even less next time. #Forecasting for Decision Making. Free webinar, Aug 11: dub.sh/IZRlhmt #BayesianStatistics #DataScience

How do you count what you can’t see? Simple models miss unobserved cases in disease outbreaks. Here's how @allendowney.bsky.social uses Bayesian inference to build a model that quantifies uncertainty 👇 allendowney.github.io/ThinkBayes2/... #Bayesian #PyMC #DataScience

How do top practitioners actually build with AI #agents in production? Chip Huyen ( @chiphuyen.bsky.social ) & Tim Hopper show Thomas Wiecki & Hugo Bowne-Anderson how they really work. Show Us Your (Agent) Skills Ep 8, tomorrow. 📅 Thurs, 7 PM ET | Live on YouTube

Show Us Your (Agent) Skills Ep 8 · Luma

What are people at the top of the game building with AI agents, and how are they doing it? Are they Claudemaxxing with eight terminals open at once? Are they…

luma.com

Everyone's talking about AI agents. Fewer talk honestly about what works when you point them at real data. So we're asking. Short survey for data scientists: what's working, what broke, what you quietly gave up on. 10-15 min, findings shared back. Survey Link: #DataScience #AIAgents

State of Agentic Data Science 2026 Survey

Made with Tally, the simplest way to create forms.

tally.so

Last day to apply for a free seat in our August cohort of Mastering #Agentic Data Science. For grad students, early-career data scientists, and anyone for whom course fees are a real barrier. Apply by end of day: dub.sh/bxCxQd8 Course details: dub.sh/41KbZYG

PyMC Labs@pymc-labs.bsky.social · last mo.

Free seats for our #Agentic Data Science course (August cohort) Live, hands-on: get #AI agents doing real analytics, catch the stats mistakes they make, keep uncertainty in your decisions. Especially for grad students, early-career & nonprofit researchers. Apply here: dub.sh/bxCxQd8

AI #Agents can build a model in minutes. They can also be confidently wrong. The skill that matters now is directing them with your context, judgment, and standards. Mastering Agentic Data Science, live cohort starting August 4th. Register 👉 dub.sh/0gFu8xo

You can read every book on #Bayesian modeling and still freeze when you point it at your own data. Chris Fonnesbeck, creator of #PyMC, on closing that gap. Applied Bayesian Modeling with Chris Fonnesbeck and @allendowney.bsky.social Starts Aug 10 Register here: dub.sh/m4qlWyb

Many strong data scientists avoid Bayesian methods, assuming they need grad-level stats and a wall of math. Hear from @allendowney.bsky.social on why that's not true: learn a few core ideas, let #PyMC handle the math, build real models from day one. live, starts Aug 10 dub.sh/2peuf1j #Bayesian

The first #datascience agent built for your team's workspace. Daimon is a purpose-built #agent for DS & has run in our team chat for over a year: 150+ MCP tools on your real stack, an isolated VM per session, right in the channel. Live webinar, July 21 dub.sh/ADOOgNo

Show Us Your (Agent) Skills: Hugo Bowne-Anderson & Thomas Wiecki host Greg Ceccarelli (SpecStory, ex-GitHub) and Han-Chung Lee (Moody's Analytics). No slides, no keynotes. Just terminals and an honest look at how they actually build with agents day-to-day. Save your spot: luma.com/y4hylb82

Show Us Your (Agent) Skills Ep 7 · Luma

What are people at the top of the game building with AI agents and how are they doing it? Are they Claudemaxxing with 8 terminals open at once? Or…

luma.com

Everyone's selling "self-improving" AI agents. Almost nobody tells you when it actually works and when it quietly makes things worse. Our new post draws the line: dub.sh/yGCfjYn #AIagents #LLM

Self-Improving AI Agents, When Skill Optimization Works

Self-improving AI agents rewrite their own skills against a benchmark. Four experiments show when automated skill optimization works — and when it fails.

pymc-labs.com

Registration is open for the August cohort of Applied #Bayesian Modeling, live with Allen Downey (@allendowney.bsky.social) and Chris Fonnesbeck (@fonnesbeck.bsky.social) 4 weeks, 8 sessions in #PyMC: priors, MCMC, hierarchical models, causal inference, time series. Enroll: dub.sh/abm-bluesky-...

We've had an open-source #AI agent living in our team chat for a year. It writes code, runs analysis, and fixes CI, all without leaving the channel called Daimon Live July 21: #agentic teammates for data science + a Daimon demo. 👉 Register here: dub.sh/adsa-bluesky #DataScience #AIagents

Free seats for our #Agentic Data Science course (August cohort) Live, hands-on: get #AI agents doing real analytics, catch the stats mistakes they make, keep uncertainty in your decisions. Especially for grad students, early-career & nonprofit researchers. Apply here: dub.sh/bxCxQd8

Before kickoff we froze a Bayesian forecast for all 48 World Cup teams. Now a state-space update in @pymc.io re-rates them off the group stage and simulates the full knockout bracket. England 14.8% to win, but the favorite still loses ~6 in 7 dub.sh/gRAdsqc #SportsAnalytics #PyMC #Worldcup2026

A new event-forecaster dpack just landed in decision-lab! Ask "when will X happen?" and it runs an ensemble of independent Bayesian forecasters in PyMC, returns one calibrated estimate, and never mistakes consensus for proof. GitHub Repo: git.new/Y678Uy3 #PyMC #Bayesian

decision-lab/decision-packs/event-forecaster at main · pymc-labs/decision-lab

Run tested, autonomous agent workflows on your data for meaningful decision-making - pymc-labs/decision-lab

git.new

You can move fast with AI agents. You can also keep your analysis trustworthy. Most teams pick one. Master Agentic Data Science (Aug cohort) is open — agents doing real analytics, not just code. Live with Hugo Bowne-Anderson, Thomas Wiecki (co-author of PyMC) and Luca Fiaschi 👉 dub.sh/rY7vXGv

📢 Just announced: Show Us Your (Agent) Skills Ep. 06. Thomas Wiecki (PyMC Labs) & Hugo Bowne-Anderson host Matt Rocklin (Dask) and Skylar Payne (Wicked Data) to show what they're building with agents. Register to join live or catch the recording: luma.com/znkfnbtz

Show Us Your (Agent) Skills Ep. 06 · Luma

What are people at the top of the game building with AI agents and how are they doing it? Are they Claudemaxxing with 8 terminals open at once? Or…

luma.com

We're open-sourcing pathmc! Structural causal models + #Bayesian estimation in one package. Define assumptions once, then estimate effects, test identifiability, run interventions, and perform sensitivity analysis. Docs: dub.sh/T0ASJt2 GitHub: dub.sh/rqiDyfg

Most optimization models assume you already know what's coming. Demand. Weather. Prices. But that forecast is almost always wrong, and your decision inherits the risk. Jun 25: PyMC Labs × FICO on optimization that stays robust when reality doesn't cooperate. Save your spot: dub.sh/0773R17

Point forecasts make optimizers confident and wrong. Jun 25, PyMC Labs × FICO: feed full posterior demand distributions into FICO Xpress, then solve with chance constraints + CVaR. Live worked example. Register: dub.sh/zhu7A4O

New PyMC model forecasts the entire 2026 World Cup group stage: scorelines and advancement odds for all 48 teams. Opening day it nailed Mexico 2-0 and Korea 2-1. England, Spain, and Brazil lead; Group D is chaos. Read the full blog here: dub.sh/hBW61BZ #PyMC #WorldCup2026

Will oil come back down? We made 5 AI analysts answer separately, then checked if they agreed. 19 years of WTI data, five independent Bayesian forecasts via Decision Lab Full case study here: dub.sh/RIUshGd #Bayesian #Forecasting

Most #MMM projects don't fail at the math. They fail at the human API. Luca Fiaschi talked with Jim Gianoglio and @simonpoulton.com on the Measure Up Podcast about how #AI agents handle the full MMM chain, and what we're building with Decision AI www.youtube.com/watch?v=AloZ...

Agentic MMM with Dr. Luca Fiaschi (PyMC Labs)

YouTube video by Measure Up | Marketing Data & Analytics Podcast

youtube.com

Production MMM breaks when influencers aren't interchangeable, campaigns shift, and channels enter mid-series. Part III of our Nürnberger case study tackles all three. 27%+ CPL reduction: dub.sh/5LLFAJ9 #MarketingMixModeling #BayesianStatistics #PyMC

Last call! Our Master Agentic Data Science course starts tomorrow. Build agent-assisted EDA, causal inference pipelines, and decision systems. Code ADS10L for 10% off. 👉 dub.link/9YG77NC #AgenticAI #DataScience #PyMC

We're sponsoring the M-Agents Hackathon at #NYCTechWeek on June 7. Go from zero to a working multi-agent AI system in one day, built with open-source tools like #PyMC and Decision Lab. Winners get a seat in our Agentic AI course. Spots are limited dub.link/ZE70nxd

@wegmansofficial.bsky.social needed to quantify cannibalization risk before opening new stores. We built a Bayesian system that answers in minutes, not weeks. Full technical walkthrough: dub.sh/1ENBxEc #RetailAnalytics #BayesianStatistics #PyMC

The radar plot is the most popular chart in football analytics. It might also be the least effective. Chris Fonnesbeck breaks down why and builds the replacement: dub.sh/BUsCcev #SportsAnalytics #DataViz #Bayesian

Data science ≠ software engineering. Automate the wrong choices, and you just get confident wrong answers faster. Tomorrow: @twiecki.bsky.social + @hugobowne.bsky.social dig into what agentic data science actually looks like. 👉Register here: luma.com/juquutuy #AgenticAI #DataScience

The Future of Agentic Data Science · Luma

AI agents have revolutionized software development. Engineers vibe code prototypes in an afternoon, refactor entire codebases in a single chat, and ship faster…

luma.com