3d ago · 6 min read · Large Language Models (LLMs) like OpenAI GPT systems or models from Google DeepMind and Meta have shown incredible capabilities in language, coding, reasoning, and content generation. Yet, despite the
Join discussion5d ago · 21 min read · Every product experimentation team doing causal inference on LLM-based features eventually hits the same wall: when the provider ships a new model version, there's no holdout. Your infrastructure team
Join discussion
May 8 · 20 min read · Causal inference for LLM-based features starts with one question editors ask before they ship anything: Did the change actually move the metric, or did the metric just move? Let's say that your team b
Join discussion
Apr 30 · 20 min read · Every product experimentation team running causal inference on LLM-based features eventually hits the same wall: when users click "Try our AI assistant," the volunteers aren't a random sample. Your pr
Join discussion
Apr 22 · 20 min read · Your team shipped an LLM-based summaries feature to wave 1 workspaces at week 20 and now the post-launch doc is due. You need a causal effect number, a specific estimate you can defend to a statistici
Join discussion
Feb 10 · 9 min read · By Hafsa Benzzi, MD, PhD Candidate in Clinical Research — Icahn School of Medicine at Mount Sinai Real-world evidence is reshaping how pharmaceutical companies develop drugs, how regulators evaluate them, and how clinicians prescribe them. In Decemb...
Join discussionSep 17, 2025 · 3 min read · Confounding can be defined as the presence of an unobserved (or ignored) covariate, which both determines the treatment allocation and the outcome to the treatment. An example that Ciaran M Gilligan Lee uses concerns vaccination during the COVID pand...
Join discussion
Sep 7, 2025 · 3 min read · Recommendation is one of the most prominent uses of machine learning in real world systems. Currently using deep learning based systems to improve recommendation is a very hot topic, and as such there is great interest in trying all sorts of fashiona...
Join discussionJan 20, 2025 · 6 min read · Why Is Causal Inference a Trend in 2025? Traditional AI (LLMs) Can Predict, Not Explain: Before: AI tools like GPT can say, "The man is likely to cheat based on patterns (correlation)." Problem: They don’t tell why cheating happens. Is it because ...
Join discussion
