Mateo Sánchez (@tembluso) header
Mateo Sánchez (@tembluso) profile picture

CS student. Building apps. Quantum SciML, PINNs, planetary science.

Joined June 2025
First physics project: 12 weeks learning CFD + Physics-Informed Neural Networks Starting from zero. Going to document what I learn and build as I figure out how to apply AI to fluid dynamics equations. Repo: github.com/tembluso/cfd-ai-j… Weekly Thread below
3
3
842
I tried to get a neural network to recover a physical constant from NASA Mars data. It converged confidently to the wrong answer. The problem: recovering the thermal diffusivity of Martian soil using PINNs.
2
8
189
The main finding isn't about model architecture or training strategy. It's about the observation operator. How you model the sensor's interaction with the physical field. Get that wrong and the PINN converges faithfully to the wrong answer. That matters more than any ML trick.
1
3
49
Week 12 recap Started writing my first scientific paper. Realized most of my results weren't polished enough. A couple of my convergence claims haven't been rigorously tested. Found even better solutions. Paper is 50%done. Extending to week 13 to finish properly.
1
81
Week 13 recap Paper is taking shape. Underestimated the amount of work it is, but just a few touches left. Final stretch.
44
Week 13 recap Paper is taking shape. Underestimated the amount of work it is, but just a few touches left. Final stretch.
1
102
Week 12 recap Started writing my first scientific paper. Realized most of my results weren't polished enough. A couple of my convergence claims haven't been rigorously tested. Found even better solutions. Paper is 50%done. Extending to week 13 to finish properly.
1
125
Week 11 recap Finished all experiments. Tried 6 different approaches to model a real-world problem. Most were dead ends but two produced results. Now polishing results and writing everything up as a proper research paper.
1
1
74
Week 10 recap Implemented the spatial representation of the sensor in the PINN RMSE dropped significantly but the recovered value was 70% larger than the published benchmark This is because the sensor itself acts as a thermal fin perturbing the temperature of the soil around.
1
1
86
I built a 2D axisymmetric PINN solving heat diffusion around the physical sensor geometry. Forward problem works, inverse problem hit a structural identifiability wall. Diagnosed exactly why and documented it.
41
Week 9 recap Ran inverse problem experiments with the PINN. It inferred a physical parameter within 7% of the published benchmark value, using a completely different method than traditional approaches.
1
1
72
However there is still an issue. The sensor averages over a surface not a point. Therefore treating it as a point measurement is physically incorrect. Next I will try to impelment the correct spatial representation.
34