On the Masters of Data, BI & AI podcast, I discussed why pricing is the most powerful data problem, what happens when AI eats software, and why intent and taste are the only competitive advantages left for humans.
Mar 31, 2026
Luca Fiaschi
A summary of my SAMI talk on synthetic consumers: using LLM-simulated buyer panels to pre-screen products, creatives, and messaging before spending real money.
Mar 25, 2026
How to combine foundational time series models with Bayesian Marketing Mix Models for scenario planning when future control variables do not exist yet.
Mar 17, 2026
PyMC Labs
A Wharton paper found that Chain-of-Thought prompting nearly matches human idea diversity. I tested what happens when you scale from 1 agent to 22. The results changed how I build AI brainstorming tools.
Mar 14, 2026
I analyzed LinkedIn data on 213 tech companies to benchmark data and AI team composition. Role mix matters more than team size.
Adapting Karpathy’s autoresearch to Bayesian causal modeling, where the treatment effect has no ground truth and the agent must optimize a proxy.
Mar 11, 2026
Engineering trust and reliability into AI agents for analytics and statistical modeling.
Feb 26, 2026
Running the IPIP Big Five personality inventory across OpenAI, Google, xAI, and open source models shows that RLHF gives each family distinct behavioral fingerprints. The patterns are real, but the statistics are trickier than they look.
Feb 23, 2026
The AI agent skills ecosystem grew explosively. Security scanning followed. But nobody is checking whether skills actually work. That’s the gap.
Feb 22, 2026
Why shipping judgment is harder than shipping code, and three pillars for building AI agents you can trust.
Feb 18, 2026
LinkedIn
Combining Optuna’s Bayesian optimization with PyMC-Marketing and CRPS to systematically tune Media Mix Model hyperparameters, instead of guessing.
Dec 20, 2025
A conversation on how synthetic consumers are reshaping product testing and market research.
Dec 1, 2025
Retailgentic
Exploring the capabilities and limitations of AI in creative evaluation.
Nov 15, 2025
Mobile Dev Memo
Why traditional statistical methods often fail in business contexts, and how Bayesian approaches can help you make better decisions under uncertainty.
Nov 12, 2025
How LLMs are changing market research and product testing through simulated consumer behavior.
Nov 1, 2025
A method enabling LLMs to generate realistic consumer survey responses, achieving 90% of human test-retest reliability.
Oct 9, 2025
arXiv
An end-to-end platform combining agentic workflows and synthetic consumers for every stage of product development.
Jun 3, 2025
My Databricks Data + AI Summit talk on how agentic AI is moving marketing science from manual processes to collaborative decision-making, with two production systems we built on Databricks.
Jun 1, 2025
How a distributed multi-agent system enables parallel evaluation of hundreds of product concepts through synthetic consumer testing.
Apr 15, 2025
MLOps Community
A systematic approach to analyzing app reviews at scale using large language models.
Dec 1, 2024
Medium
A practical discussion on the state of Bayesian MMMs what works, what doesn't, and how companies like HelloFresh use advanced modeling to guide budget allocation, forecasting, and long-term customer value decisions.
Nov 1, 2022
Building Morpheus, a machine learning algorithm for weekly customer-level lifetime value predictions.
Dec 7, 2020
HelloFresh Tech
How HelloFresh built a Bayesian media mix model with PyMC3 to optimize marketing spend allocation.
Aug 24, 2020
Rethinking personalization beyond traditional recommendation systems at HelloFresh.
Jan 1, 2020
