What Ishaan and Vedant Learned This Summer

Typical Street View coverage of La Paz

Antrum Platonicum by Jan Saenredam (after Cornelis Cornelisz van Haarlem)

As the first of our interns wrap up and prepare for the fall semester, we wanted to ask them a few questions to help them reflect on their time at Rowan and share learnings with the world.

Today, both Ishaan and Vedant are concluding their internships with Rowan. We're excited to share more of what they've worked on soon. In the meantime, we hope you enjoy their responses to a few questions:

Ishaan's Answers

What are five things you learned during your internship at Rowan?

  1. Chemistry and biology software packages are annoying to work with and are largely incompatible with each other. Moreover, because they aren't super widely used, LLMs don't do a great job of suggesting good fixes when one runs into library-specific bugs.

  2. Blogs are an awesome source of information for both technical and nontechnical content.

  3. Staying up to date with the literature in a field is really helpful in giving you a clear picture of the field's current state. Transitioning from reading an occasional paper to reading a paper daily feels like being freed from the cave and going outside.

  4. Neural network potentials, as great as they are for small molecule property prediction, do a very poor job of predicting protein–ligand interaction energies. This is even the case for the Meta models trained on the OMol25 dataset. [For more on this, see Ishaan's work benchmarking protein–ligand interaction energies.]

  5. "416" is a Toronto phone number prefix. "816" is Kansas City. A big city in the mountains in South America is probably either La Paz or Quito.

Typical Street View coverage of La Paz

Typical Street View coverage of La Paz

What's one thing you're hoping to learn more about following your time at Rowan?

Definitely improving MD [molecular dynamics] simulations. I think there are a lot of cool ideas in this space that are largely good "in theory" but haven't been implemented effectively for easy usage. I want to work on/study different ways to run, say, mid-length MD simulations (e.g. 40 ns) more quickly than existing approaches.

Ishaan is returning to Brown University this fall, where he's studying applied math, CS, and physics. This summer, he also released a preprint on cavity polaritons.

Vedant's Answers

What are five things you learned during your internship at Rowan?

  1. Applied research in industry requires a faster pace and narrower focus than academic research (no getting nerd sniped).

  2. To work in an interdisciplinary environment with computer scientists, chemists, biologists, etc. you need to be able to both absorb and filter a lot of information at once and be able to explain what you're doing in a way that is understandable (minimize or always define jargon).

  3. Similarly, to give a research talk to a group with varying backgrounds, you should focus on building intuition while not sacrificing rigor, ensuring that everyone can get something out of your talk. It should be informative, succinct, and tailored to the audience.

  4. ML [machine learning] has a lot of promise in scientific fields but is not a solution for every problem (as much as that realization may "pain" me and other ML researchers). Like any tool or framework, context is key.

  5. There's a gap between ML papers reporting new methods and what scientists actually care about. Usefully addressing this gap requires rigorously testing methods, looking at data splitting, benchmarking, and carefully examining any potential data leakage.

What's one thing you're hoping to learn more about following your time at Rowan?

The intersection between the natural sciences (specifically physics/chemistry) and deep learning methods; it seems like there are a lot of interesting problems and that improving my overall scientific understanding would help.

Vedant is returning to Northeastern University this fall, where he's a master's student working on parameter space symmetries in transformers.

Banner background image

What to Read Next

Boltz-2 Constraints, Implicit Solvent for NNPs, and More

Boltz-2 Constraints, Implicit Solvent for NNPs, and More

new terms of service; comparing IRCs and conformer searches; contact and pocket constraints for Boltz-2; MOL2 download; implicit-solvent NNPs; draft workflows; optimizing docking efficiency
Sep 22, 2025 · Corin Wagen, Ari Wagen, Jonathon Vandezande, Eli Mann, and Spencer Schneider
Controlling the Speed of Rowan's Docking

Controlling the Speed of Rowan's Docking

Some notes on how docking can be tuned for different applications.
Sep 22, 2025 · Corin Wagen
Studying Scaling in Electron-Affinity Predictions

Studying Scaling in Electron-Affinity Predictions

Testing low-cost computational methods to see if they get the expected scaling effects right.
Sep 10, 2025 · Corin Wagen
Open-Source Projects We Wish Existed

Open-Source Projects We Wish Existed

The lacunæ we've identified in computational chemistry and suggestions for future work.
Sep 9, 2025 · Corin Wagen, Jonathon Vandezande, Ari Wagen, and Eli Mann
How to Make a Great Open-Source Scientific Project

How to Make a Great Open-Source Scientific Project

Guidelines for building great open-source scientific-software projects.
Sep 9, 2025 · Jonathon Vandezande
ML Models for Aqueous Solubility, NNP-Predicted Redox Potentials, and More

ML Models for Aqueous Solubility, NNP-Predicted Redox Potentials, and More

the promise & peril of solubility prediction; our approach and models; pH-dependent solubility; testing NNPs for redox potentials; benchmarking opt. methods + NNPs; an FSM case study; intern farewell
Sep 5, 2025 · Eli Mann, Corin Wagen, and Ari Wagen
Machine-Learning Methods for pH-Dependent Aqueous-Solubility Prediction

Machine-Learning Methods for pH-Dependent Aqueous-Solubility Prediction

Prediction of aqueous solubility for unseen organic molecules remains an outstanding and important challenge in computational drug design.
Sep 5, 2025 · Elias L. Mann, Corin C. Wagen
What Isaiah and Sawyer Learned This Summer

What Isaiah and Sawyer Learned This Summer

Reflections from our other two interns on their time at Rowan and what they learned.
Sep 5, 2025 · Isaiah Sippel and Sawyer VanZanten
Benchmarking OMol25-Trained Models on Experimental Reduction-Potential and Electron-Affinity Data

Benchmarking OMol25-Trained Models on Experimental Reduction-Potential and Electron-Affinity Data

We evaluate the ability of neural network potentials (NNPs) trained on OMol25 to predict experimental reduction-potential and electron-affinity values for a variety of main-group and organometallic species.
Sep 4, 2025 · Sawyer VanZanten, Corin C. Wagen
Which Optimizer Should You Use With NNPs?

Which Optimizer Should You Use With NNPs?

The results of optimizing 25 drug-like molecules with each combination of four optimizers (Sella, geomeTRIC, and ASE's implementations of FIRE and L-BFGS) and four NNPs (OrbMol, OMol25's eSEN Conserving Small, AIMNet2, and Egret-1) & GFN2-xTB.
Sep 4, 2025 · Ari Wagen and Corin Wagen