Austin Tripp's website

☀️ High confidence 🤖 AI-assisted 💎 High effort 💡 High originality 🌱 New During a discussion1 with Claude, it raised the idea of using sketching (random linear projections) to compress molecular fingerprints.

☀️ High confidence 💎 High effort 🌱 New My content standards for this blog have changed over time.

⛅ Medium confidence 🤖 AI co-written 🔧 Moderate effort 💡 High originality 🌱 New Preamble This post was the result of me putting Claude Fable to the test in early July when it was made available to the public again and was included in my personal Claude Pro…

🌱 New I recently read the paper1 “Grand Challenges for Predictive Modeling in Small Molecule Drug Discovery”, which aims to survey all the challenges in the field I work in.

☀️ High confidence 🤖 AI-assisted 💎 High effort 🧩 Medium originality 🌱 New I recently found myself wanting to use the Vecchia GP approximation for a problem at work where low-rank approximation methods were a poor fit.

⛅ Medium confidence 💎 High effort 💡 High originality I just wrote an article How AI Experts Can Drive Better Collaboration in Drug Discovery on Valence’s substack, reflecting on some experiences I’ve had at Valence/Recursion.

After switching my website’s build from Jekyll to Nikola in 2024 (more details here) I’ve decided to switch again to Quarto.

🪞 Low originality A colleague recently introduced me to Quarto and I can’t believe I didn’t know about it sooner!

✏️ Quick post Don’t write your paper for somebody to read it, write for somebody to skim it.

I’ve been a somewhat late adopter to coding agents: until ~1 month ago I was mainly using cursor and asking it to suggest local edits, or occasionally bigger refactors.

I recently ran an LLM deep research query about large matrices of Tanimoto coefficients,1 and one of its references was the paper “Bandwidth adjustable Tanimoto kernel: a smooth alternative to the Gaussian kernel” (https://doi.org/10.1007/s41237-026-00296…

This is a re-post from my book summaries repo, which is my main home for book summaries.

⛅ Medium confidence 🔧 Moderate effort 🧩 Medium originality I think the Bayesian optimization (BO) research community needs to change its evaluation practices.

⛅ Medium confidence 🔧 Moderate effort 🧩 Medium originality In a recent blog post (link) I described my “model-centric” view of Bayesian optimization (BO), essentially arguing that the model is the most important component of BO and BO users (and researche…

☀️ High confidence 🔧 Moderate effort 🪞 Low originality The most common kind of Gaussian process (GP) model is: Essentially, the function of interest is sampled from a Gaussian process, and observations are noisy versions of the function values with Gaussi…

Cross-posting my summary of the book “Talent: How to Identify Energizers, Creatives, and Winners Around the World” from my GitHub: link.

⛅ Medium confidence 🔧 Moderate effort 🧩 Medium originality I think the Bayesian optimization (BO) community is vastly underselling one of BO’s most practically appealing features: modularity.

☀️ High confidence 🔧 Moderate effort 🧩 Medium originality A lot of Bayesian optimization (BO) papers include experiments on “toy” functions.

☀️ High confidence 💎 High effort 💡 High originality Bayesian decision theory is one of the best justifications for BO- particularly for myopic acquisition functions like expected improvement.

☀️ High confidence 💎 High effort 🧩 Medium originality A lot of conversations at NeurIPS this year made me think that I view the role of the surrogate model in Bayesian optimization a bit differently than many other researchers in the field, and this profo…