Killed the code, Killed by the interview.
AI-Atlas score
| Machine Learning | 69 |
|---|---|
| Computer Vision | 51 |
| Data Science | 48 |
| NLP & LLMs | 59 |
| AI Agents | 36 |
| MLOps | 45 |
Each node is one domain, scored only from evidence in these public repositories. Select one to see why.
Deep, repeated, well-engineered work in this domain.
37 public repositories → 13 with AI work, evaluated. The other 24 (repositories without AI code, forks and empty repositories) are not scored and have no effect on the result. How repositories are selected
Ordered by how much each project adds to the Atlas score, not by its own score. A project with evidence in several domains can rank above one with a slightly higher project score. Choose a domain to focus on the projects with meaningful evidence there.
23% of Atlas Score
Demolytics is an end-to-end machine learning application for sentiment analysis of Indonesian YouTube comments using transformer models (IndoBERT and IndoBERTweet) and a custom ensemble approach. It includes data preprocessing, model training with class imbalance handling, evaluation, and deployment via Docker.
20% of Atlas Score
The repository implements Random Forest and Isolation Forest from scratch using JAX, including data preprocessing, multi‑GPU sharding, stump‑based and recursive tree ensembles, and evaluation against a heuristic label, with saved model pickles for reuse.
19% of Atlas Score
This repository provides a vendor-agnostic PPO training pipeline for a custom hexapod robot, supporting both MuJoCo (PyTorch) and Brax (JAX) backends with unified policy export, configuration-driven experiments, and Dockerized deployment.
13% of Atlas Score
This repository provides an end-to-end pipeline for training and evaluating image classification models (MobileNetV3-Small and ResNet-18) and an object detection model (YOLOv8n) on a medical pill dataset, with a focus on comparing the two classification models.
7% of Atlas Score
This repository provides an end-to-end pipeline for sentiment analysis in Indonesian educational social media text, including data preprocessing, translation, and model training using transformer-based architectures, with containerization for environment reproducibility.
6% of Atlas Score
This repository provides a Jupyter notebook that loads historical gold price and macroeconomic data, builds and trains an LSTM model using PyTorch for price forecasting, and evaluates performance via time‑series cross‑validation. It includes Docker configuration for containerized execution.
