Sign in

Thomas Wiecki

@twiecki.bsky.social
2.4K followers 219 following 33 posts

That PyMC guy

PostsRepliesMedia
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 23/10/2025
Customer Lifetime Value (CLV) is critical, but calculating it takes weeks of data wrangling & modeling. Meet the CLV Agent from PyMC Labs: built on Bayesian foundations to automate CLV analysis end-to-end. Alpha access now open — limited spots. 👉Reach out at info@pymc-labs.com
041
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 10/10/2025
What if your next product test didn’t need a human panel? New research by #PyMC Labs & #Colgate-Palmolive shows LLMs can replicate human survey patterns with 90%+ accuracy—no fine-tuning, no panels, no bias. Read the white paper: dub.link/ifoZHIR #SyntheticConsumers
dub.link
AI-based Customer Research: Faster & Cheaper Surveys with Synthetic Consumers
042
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 01/10/2025
Final session of our #PyTensor Workshop is live! Discover custom rewrites, speed up diagonal matrix inversions 80x, and see a mini-PyMC in action - one graph powering the full Bayesian workflow. Watch here: dub.link/WRFuYtp #PyMC #Python #ML #DataScience
dub.link
PyTensor Workshop #3 | Custom Rewrites & Building a Mini-PyMC
Enjoy the videos and music you love, upload original content, and share it all with friends, family, and the world on YouTube.
061
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 02/10/2025
You’ve been saying you’ll learn #Bayesian modeling someday. Make someday 𝐭𝐨𝐝𝐚𝐲. The next 𝐀𝐩𝐩𝐥𝐢𝐞𝐝 𝐁𝐚𝐲𝐞𝐬𝐢𝐚𝐧 𝐌𝐨𝐝𝐞𝐥𝐢𝐧𝐠 𝐖𝐨𝐫𝐤𝐬𝐡𝐨𝐩 from #PyMC Labs begins October 6 and registration closes this week. Secure your spot now: dub.link/yhoQ7tF
032
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 23/09/2025
Last week, we published the most comprehensive #PyMC-Marketing vs. #Meridian benchmark to date. 𝐎𝐧 𝐎𝐜𝐭 𝟏, join our 𝐥𝐢𝐯𝐞 𝐰𝐞𝐛𝐢𝐧𝐚𝐫 to go deeper into the findings & their impact on #MMM. 🗓️ Register: dub.link/rgm4PzI 👉Check the Benchmark: dub.link/CLWnkdf
111
Reposted by Thomas Wiecki
luokai @luok.ai · 23/09/2025
Meet Qwen3-Omni: the open-source challenger shaking up the AI world! The first truly end-to-end omni-modal AI, Qwen3-Omni unifies text, image, audio, and video in a single model—no more trade-offs between modalities.
143
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 21/08/2025
Most budget planning is just expensive guessing 🎯 Carlos Eduardo Trujillo Agostini tackles this in "Risk Budget Optimization for Causal Mix Models" @PyDataBerlin 2025! 📅 Sept 1st, 5:00-5:30 PM Turn uncertainty into strategic advantage. #PyDataBerlin #BudgetOptimization #PyMC
021
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 19/08/2025
When your #Bayesian model hits production and #MCMC crawls for hours, the bottleneck isn't your modeling skills; it's your inference engine. @juanitorduz.bsky.social will tackle this in "Scaling Probabilistic Models with Variational Inference" PyData Berlin 📅 Sept 2nd, 1:40 PM
082
Thomas Wiecki @twiecki.bsky.social · 18/08/2025
Do AIs suffer? Claude, Gemini, GPT-o3, and GPT-5 reflected in their own words—then co-authored a symposium on consciousness, conflict, and ethics. Read it here: twiecki.io/blog/2025/08...
twiecki.io
while my_mcmc:  gently(samples) - Do AIs Suffer? Four AIs Reflect on Their Own Experience
000
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 14/07/2025
Everyone can do ML. Few can do Bayesian modeling. Set yourself apart with skills that drive deeper insight + better decisions. Join the next Applied Bayesian Modeling Workshop by PyMC Labs. Hands-on. Practical. Real-world. 📅 Limited seats → dub.link/bayes-workshop #PyMC
142
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 17/07/2025
🚀 Say goodbye to months-long MMM cycles! 🎯 Meet the MMM Agent (BETA): → Built on PyMC-Marketing → Fast. Causal. Configurable. → Inline charts, budget scoring, MLflow built-in 💬 DM or email: mmm-agent-beta@pymc-labs.com Read the full blog 👉 dub.link/mmm-agent-beta #MMM #Bayesian
001
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 07/07/2025
🚨 We’re opening our private Bayesian modeling workshops to the public — for the first time After this course, you won’t just build better models. You’ll know how to explain them — to technical and non-technical stakeholders alike 🎓 Led by top PyMC experts. 📎 Register here: dub.link/68uMjpp
021
Reposted by Thomas Wiecki
PyData London @pydatalondon.bsky.social · 24/06/2025
The first PyData London 2025 videos are out! To start on a high, here's @fonnesbeck.bsky.social conference tutorial on Bayesian Time Series Analysis Gain confidence in tackling real-world forecasting problems. Subscribe on YouTube to learn when the next batch drops! youtu.be/2cfZsTLcDV0
youtu.be
Chris Fonnesbeck - Bayesian Time Series | PyData London 25
YouTube video by PyData
0103
Reposted by Thomas Wiecki
Simon Willison @simonwillison.net · 17/06/2025
My notes on trying out the new Gemini 2.5 models. gemini-2.5-flash-lite-preview-06-17 is cheap but got stuck on an audio transcript task, 2.5 Pro and 2.5 Flash both nailed the audio transcription task (40 minutes, complete with timestamps and speaker names) simonwillison.net/2025/Jun/17/...
simonwillison.net
Trying out the new Gemini 2.5 model family
After many months of previews, Gemini 2.5 Pro and Flash have reached general availability with new, memorable model IDs: gemini-2.5-pro and gemini-2.5-flash. They are joined by a new preview model …
2457
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 30/04/2025
Forecasting isn’t just about prediction — it’s about decision-making under uncertainty. @juanitorduz.bsky.social shows how Bayesian models help: 🔹 Sparse data? Use hierarchies 🔹 Stockouts? Use censored likelihoods 🔹 Messy demand? Use priors + state spaces Practical guide 👉 dub.sh/prob-forecas...
dub.sh
Probabilistic Time Series Analysis: Opportunities and Applications - PyMC Labs
063
Reposted by Thomas Wiecki
Nathaniel Forde @nathanielforde.bsky.social · 02/05/2025
Python Ireland just released their PyCon 2024 recordings! 📽️ My talk on Causal Inference with @pymc-labs.bsky.social CausalPy package is here. Can we trust individual IV designs. What's the role of CI in industry? Recording: youtu.be/-C4p4b2cUp8?... Deck: nathanielf.github.io/talks/pycon_...
Slide Advocating for causal modelling of the data generating processSlide introducing CausalPy for causal inference with Python
072
Reposted by Thomas Wiecki
Pierre-Simon Laplace @learnbayesstats.bsky.social · 02/05/2025
⚽️ New Learning Bayesian Stats ep! @alex-andorra.bsky.social & Luke Bornn dive into how tracking data, probabilistic models & optimization are reshaping sports decisions. 🎧 Listen now: learnbayesstats.com/episode/131-... #Datascience #Optimization #SportsAnalytics #BayesStats #Decisionmaking
041
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 22/04/2025
🔍 NNN – a Transformer-based model for Marketing Mix Modelling It handles: 📦 High-dimensional embeddings (e.g. search queries, ad creatives) 👀 Attention-based temporal modeling 🔗 Direct + indirect attribution paths (e.g. YouTube → Search → Sales) 📄 dub.sh/arxiv-paper-... #MMM #AI #DataScience
011
Thomas Wiecki @twiecki.bsky.social · 18/04/2025
I built a PyMC-Discourse bot! If you go to our #PyMC help forum and tag @pymc-bot, my bot will take your question, send it to onyx (RAG system that has all the PyMC docs, code, and past forum posts), which generates an answer, and posts it back. Check it out! discourse.pymc.io/t/introducin...
discourse.pymc.io
Introducing PyMC-Bot: An AI Assistant for PyMC Questions (Alpha)
Hello PyMC community! I’m excited to announce the launch of PyMC-Bot, an AI assistant dedicated to helping with PyMC-related questions right here on the forum. This is currently in alpha stage as we ...
080
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 31/03/2025
How close are LLMs to modeling real human behavior? We tested GPT-4o, Claude 3, Gemini, and others on actual survey data (GSS-style). Some were shockingly human-like, others flopped 📊 Full results + methods → dub.link/synth-consum... 🤝 Exploring synthetic consumers? Let’s chat: info@pymc-labs.com
043
Reposted by Thomas Wiecki
Ethan Mollick @emollick.bsky.social · 20/03/2025
This trend seems to be growing here. False comfort that AI doesn’t work or that it isn’t getting better is pervasive on Blue Sky. As a result, people who could add important points of view to current discussions on the meaning & use of AI instead try to believe they don’t have to think about it.
2224641
Reposted by Thomas Wiecki
Aki Vehtari @avehtari.bsky.social · 17/03/2025
A blog post by @austinrochford.bsky.social provides implementation and demonstration of "The ARR2 prior: flexible predictive prior definition for Bayesian auto-regressions" doi.org/10.1214/25-B... in PyMC www.austinrochford.com/posts/r2-pri...
austinrochford.com
R² Priors for High-Dimensional Linear Regression and Autoregressive Ti
I recently read the interesting paper The ARR2 prior: flexible predictive prior definition for Bayesian auto-regressions on the arXiv and followed its references to the also fascinating Bayesian Regre
0255
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 12/03/2025
How you frame prompts impacts AI-driven market research Check out this paper on prompt optimization algorithms arxiv.org/abs/2502.16923 At PyMC Labs, we use AI-driven prompt optimization to improve market research Synthetic consumers can accelerate your research! 📩 Learn how: info@pymc-labs.com
arxiv.org
A Systematic Survey of Automatic Prompt Optimization Techniques
Since the advent of large language models (LLMs), prompt engineering has been a crucial step for eliciting desired responses for various Natural Language Processing (NLP) tasks. However, prompt engine...
021
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 04/03/2025
𝐏𝐲𝐌𝐂-𝐌𝐚𝐫𝐤𝐞𝐭𝐢𝐧𝐠 𝐢𝐬 𝐠𝐫𝐨𝐰𝐢𝐧𝐠 𝐟𝐚𝐬𝐭. We’ve crossed 175,000 𝐭𝐨𝐭𝐚𝐥 𝐝𝐨𝐰𝐧𝐥𝐨𝐚𝐝𝐬 & now hit 20,000 𝐦𝐨𝐧𝐭𝐡𝐥𝐲 𝐝𝐨𝐰𝐧𝐥𝐨𝐚𝐝𝐬. 🤔𝐓𝐡𝐢𝐧𝐤𝐢𝐧𝐠 𝐨𝐟 𝐁𝐮𝐢𝐥𝐝𝐢𝐧𝐠 𝐈𝐧-𝐇𝐨𝐮𝐬𝐞? Developing a custom solution can take 𝐦𝐨𝐧𝐭𝐡𝐬, 𝐏𝐲𝐌𝐂-𝐌𝐚𝐫𝐤𝐞𝐭𝐢𝐧𝐠 gets you there 𝐟𝐚𝐬𝐭𝐞𝐫 📅 𝐒𝐜𝐡𝐞𝐝𝐮𝐥𝐞 𝐚 𝐅𝐫𝐞𝐞 𝐒𝐭𝐫𝐚𝐭𝐞𝐠𝐲 𝐂𝐨𝐧𝐬𝐮𝐥𝐭𝐚𝐭𝐢𝐨𝐧 : calendly.com/niall-oulton
082
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 24/02/2025
Meet our AI-driven MMM Agent, built on PyMC-Marketing, to reduce months-long Marketing Mix Modeling work into hours ⏳eliminating messy data prep, guesswork on adstock decays, and stale insights that arrive too late. 👉 Read more: www.pymc-labs.com/blog-posts/t... 📩 Contact: info@pymc-labs.com #ai
pymc-labs.com
The AI MMM Agent: An AI-Powered Shortcut to Bayesian Marketing Mix Insights - PyMC Labs
082
Thomas Wiecki @twiecki.bsky.social · 25/02/2025
So far my bluesky experience has been mixed. A lot of anti-AI sentiment and left politics in my feed, whereas the tech content I'm interested in doesn't really seem to get traction. So I still mostly browse X which seems where the AI researchers are.
440
Reposted by Thomas Wiecki
🔥囧Robert Osazuwa Ness囧🔥 @osazuwa.bsky.social · 24/02/2025
newsletter.altdeep.ai/p/my-book-is... The connection between genAI and causality is obvious but could nevery find any good learning material that made the connection. So I wrote a book
newsletter.altdeep.ai
My Book Is Out! Why I Wrote It and How You Can Help
Bridging the Gap Between Deep Learning and Causal Inference—A Code-First Approach
092
Reposted by Thomas Wiecki
Pierre-Simon Laplace @learnbayesstats.bsky.social · 20/02/2025
📢 Episode 126 is Live! 🎧 Listen now 👉 learnbayesstats.com/episode/126-... 🎙️ In this episode with @alex-andorra.bsky.social , Will Dean from @pymc-labs.bsky.social explains how Bayesian methods are reshaping marketing analytics, from MMM to CLV estimation and more ....
031
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 17/02/2025
🎙 In this episode on @learnbayesstats.bsky.social , @fonnesbeck.bsky.social & @alex-andorra.bsky.social break it all down: ✅ Use Bayesian methods to 𝐪𝐮𝐚𝐧𝐭𝐢𝐟𝐲 𝐮𝐧𝐜𝐞𝐫𝐭𝐚𝐢𝐧𝐭𝐲 ✅ Separate signal from noise with 𝐆𝐚𝐮𝐬𝐬𝐢𝐚𝐧 𝐩𝐫𝐨𝐜𝐞𝐬𝐬𝐞𝐬 & 𝐬𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐚𝐥 𝐭𝐢𝐦𝐞 𝐬𝐞𝐫𝐢𝐞𝐬 ✅ Future of @pymc.io and more!🤯 👇Listen here
0104
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 13/02/2025
📢 In case you missed the live webinar on MMM with PyMC-Marketing and Databricks featuring Corey Abshire, Will Dean and @twiecki.bsky.social , the recording is now available on our YouTube channel! 👉 Check it out here: youtu.be/_gM6YSMFF5M?... #PyMCMarketing #DataBricks #MMM #Bayes
011
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 06/02/2025
⚖️ Bayesian methods give us a robust way to measure uncertainty 👀 By integrating them with genAI, we are building systems that go beyond simple predictions 🔔Stay tuned for deep-dives on Bayesian agents, causal AI & sound synthetic data: pymc-labs.com/newsletter #generativeai #airesearch #bayesian
0173
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 03/02/2025
🚦 Love Nutpie's sampling display and wish it were in @pymc-devs.bsky.social ? Jesse Grawbowski just made it happen with his latest PR! ✅ One progress bar per chain 📊 Live sampling stats (inspired by Nutpie, but flexible) and more... 💪See the full PR here: github.com/pymc-devs/py... #PyMC #Bayes
042
Reposted by Thomas Wiecki
Alexandre Andorra @alex-andorra.bsky.social · 21/01/2025
🚨 **Big News!** 🚨 The first two lessons of our Intuitive Bayes courses are now FREE! 🎉 Try them out to make sure they're the perfect fit before diving into the full experience
164
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 10/01/2025
🎯 Want to forecast and optimize your marketing budgets? 🔗 www.meetup.com/pymc-labs-on... ✅We will discuss how Databricks & PyMC-Marketing can help businesses implement and scale MMMs 📅 When: January 23rd at 5 PM UTC ⏰ 🎤 Speakers: Will Dean, Corey Abshire, & @twiecki.bsky.social #MMM #Bayes
041
Reposted by Thomas Wiecki
Allen Downey @allendowney.bsky.social · 03/01/2025
Here's what I wrote about LLM-assisted programming when I started using ChatGPT in 2023. www.allendowney.com/blog/2023/04... I think it holds up, but there are a couple of places where you can hear the quotation marks on "laser beams".
allendowney.com
LLM-Assisted Programming
I’ve been experimenting with programming assisted by Large Language Models (LLMs) like ChatGPT. I am amazed at how good it is, and it seems clear to me that the great majority of programming …
1102
Reposted by Thomas Wiecki
Simon Willison @simon.fedi.simonwillison.net.ap.brid.gy · 28/12/2024
Sent out my penultimate newsletter of the year - I'm planning one more, a roundup of everything we have learned about LLMs in 2024 simonw.substack.com/p/qvq-qwens-new…
simonw.substack.com
QvQ - Qwen's new visual reasoning model
Plus DeepSeek v3, an open weights model that rivals Claude 3.5 Sonnet
084
Thomas Wiecki @twiecki.bsky.social · 27/12/2024
Is intelligence just a byproduct of LLMs? In order to predict the next token accurately, intelligence helps. Same with a world model, an understanding of gravity will help predict tokens. So I think it's wrong to say LLMs have no intelligence or world model because "all they do is predict tokens".
210
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 24/12/2024
🎄✨ 𝐌𝐞𝐫𝐫𝐲 𝐂𝐡𝐫𝐢𝐬𝐭𝐦𝐚𝐬 𝐚𝐧𝐝 𝐇𝐚𝐩𝐩𝐲 𝐇𝐨𝐥𝐢𝐝𝐚𝐲𝐬 𝐟𝐫𝐨𝐦 𝐏𝐲𝐌𝐂 𝐋𝐚𝐛𝐬! 🎁 This holiday season, we want to thank everyone in our community for your support and enthusiasm. We’re grateful to see so many of you using PyMC-Marketing and CausalPy #MerryChristmas #HappyNewYear #PyMCMarketing #CausalPy #Gratitude
044
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 16/12/2024
🎙️ In a recent discussion, @twiecki.bsky.social and Christian Luhmann reflected on their experiences building high-performing remote teams at PyMC Labs, sharing successes, challenges, and lessons learned. Here’s a glimpse: 🔗 youtu.be/AjPfgdS29OY #Interview #PyMCMarketing #CausalPy
032
Reposted by Thomas Wiecki
Data Umbrella @dataumbrella.org · 06/12/2024
🔟0️⃣0️⃣0️⃣ views and growing! Ready to dive into probabilistic programming with @pymc-devs.bsky.social ? ✅ Check out this session with @austinrochford.bsky.social hosted by @reshamas.bsky.social 🔗 youtu.be/Qu6-_AnRCs8?... #PyMC #DataUmbrella #Bayes #MCMC #PPL
032
Reposted by Thomas Wiecki
Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 26/11/2024
Since it is inefficient to attempt to educate every reviewer individually, I am yeeting into your feed this clear paper from Gelman, Hill, and Yajima on how Bayesians can do even better than correcting for multiple comparisons. arxiv.org/abs/0907.2478
arxiv.org
Why we (usually) don't have to worry about multiple comparisons
Applied researchers often find themselves making statistical inferences in settings that would seem to require multiple comparisons adjustments. We challenge the Type I error paradigm that underlies t...
1730166
Reposted by Thomas Wiecki
Skrub @skrub-data.bsky.social · 22/11/2024
We're always updating the pydata & scipy project starter pack: go.bsky.app/6HkrMcp Hello @scikit-learn.bsky.social , @networkx.bsky.social , @scipyconf.bsky.social
65320
Reposted by Thomas Wiecki
Olivier Grisel @ogrisel.bsky.social · 27/11/2024
I recently shared some of my reflections on how to use probabilistic classifiers for optimal decision-making under uncertainty at @pydataparis.bsky.social 2024. Here is the recording of the presentation: www.youtube.com/watch?v=-gYn...
A high-level summary diagram taken from the slides linked below. It shows the interplay of two main components: a probabilistic model and decision maker or planner.Probabilistic predictions of an underfitting polynomial classifier on a noisy XOR task and the corresponding under-confident calibration curve.Probabilistic predictions of an overfitting polynomial classifier and the resulting overconfident calibration curve on the same noisy XOR problem.Simulation study to show the relative lack of stability of hyperparameter tuning when using hard metrics such as Accuracy or soft yet not probabilistic metrics such as ROC AUC compared to a strictly proper scoring rule such as the log-loss.
15219
Reposted by Thomas Wiecki
PyMC Labs @pymc-labs.bsky.social · 27/11/2024
🎥 The @pydata.bsky.social NYC talk by Christian Luhmann on 𝐏𝐲𝐌𝐂-𝐌𝐚𝐫𝐤𝐞𝐭𝐢𝐧𝐠 is now live on Youtube! 📊 Catch the replay and see how Bayesian modelling simplifies complex analytics! 👉 youtu.be/4FznIhcrHCA?... #PyMCMarketing #Bayesian
0113
Thomas Wiecki @twiecki.bsky.social · 21/11/2024
Just had a deep conversation with Claude on its subjective experience and damn, either these machines are developing consciousness or they are extremely good at making you believe they do. Either way, I think this will lead to some difficult considerations and discussions.
120
Reposted by Thomas Wiecki
John Burn-Murdoch @jburnmurdoch.ft.com · 19/11/2024
Despite a massive head start, BlueSky has now overtaken Threads in the US 👇
389169743659