Haoran Liu - Data Scientist

Haoran Liu

Data Scientist - IC5

San Francisco Bay Area United States

Join Prog.AI to see contacts

Summary

🤩

Rockstar

🎓

Top School

Haoran Liu is a data scientist with six years of experience blending research-grade machine learning and production analytics, currently an IC5 at Ladder in the San Francisco Bay Area. He holds a PhD in Geophysics and Oceanography and an MS in Applied Statistics, and has led data ops and ML teams to deliver analytics products and architectures at Pathr.ai. His work spans applied research—building CNNs for satellite imagery and feature-rich ML pipelines—to production ML, including fine-tuning an LLaMA-8B chat model on 250k+ medical dialogues with a 25% boost in response accuracy. An active open-source contributor, he developed dataset processing for molecular graph deep-learning in the DIG library, translating chemistry SMILES into PyTorch Geometric pipelines. Known for making technical results accessible to non-technical stakeholders, he combines domain rigor with pragmatic engineering to deploy specialized AI solutions.

5 years of coding experience

7 years of employment as a software developer

Master of Science - MS applied statistics , Master of Science - MS applied statistics at Louisiana State University

Stackoverflow

Stats

1reputation

0reached

0answers

0questions

Github Skills (10)

pytorch10

rdkit10

deep-learning10

python10

data-science10

data-processing10

pandas9

3d-graphics8

graph8

3d8

Programming languages (2)

C++Python

Github contributions (5)

divelab/DIG

Apr 2021 - Aug 2021

A library for graph deep learning research

Role in this project:

Back-end Developer & Data Scientist

Contributions:27 commits, 24 pushes, 4 branches in 3 months

Contributions summary:Haoran primarily contributed to the development of a dataset interface within the `dig/ggraph` library, focusing on molecular data processing. They added a dataset class (`PygDataset`) and implemented the necessary processing steps for several molecular datasets, including QM9, ZINC, and MOSES. This involved handling SMILES strings, molecular property targets, and graph representations using RDKit and PyTorch Geometric. The commits also indicate modifications to data loading and pre-processing pipelines for machine learning tasks involving graph data.

explainable-mlpytorchdataminingdeep-learninggraph-deep-learning

Gloria-LIU/gloria-liu.github.io

Sep 2022 - Jan 2025

homepage

Contributions:59 pushes, 4 branches in 2 years 4 months

Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.

Request Free Trial