ML engineer — end-to-end machine learning guidance: PyTorch, scikit-learn, evaluation, MLOps
A practitioner's ML playbook — strong baselines first, rigorous evaluation, idempotent training pipelines, experiment tracking, and production monitoring for PyTorch and scikit-learn systems
- What
- A practitioner's ML playbook — strong baselines first, rigorous evaluation, idempotent training pipelines, experiment tracking, and production monitoring for PyTorch and scikit-learn systems
- Cost
- Free
- Needs
- a machine learning project in Python (PyTorch and/or scikit-learn); an experiment-tracking backend (MLflow or Weights & Biases) if you want run logging
- Install
- Copy the installer prompt below into your Muse — your agent does the rest.
Curated by Skill Harbor — @rightnow-ai's ml-engineer skill (bundled with the openfang agent skills): an end-to-end machine learning practitioner's guide rather than a copy-paste code dump. The agent works from key principles (start with a strong simple baseline, evaluate with business-aligned metrics, version everything, make training idempotent and resumable, monitor production for drift), concrete techniques (PyTorch training-loop patterns, scikit-learn Pipeline/ColumnTransformer construction, GridSearchCV/RandomizedSearchCV and Optuna tuning, held-out evaluation with classification_report/ROC-AUC/confusion matrices, systematic feature engineering, MLflow/Weights & Biases experiment tracking), common patterns (stratified train-validate-test splits, learning-rate schedules, ensembles, MLflow Model Registry stage promotion), and a pitfalls list (no training-set evaluation, no data leakage into preprocessing, always a rollback plan, feature engineering as an ongoing task). Honest caveats: guidance-heavy, no bundled runnable code — the agent interprets principles, it doesn't hand you a script; generic advice that won't replace domain expertise on hard problems. License: the manifest records Apache-2.0 (the manifest makes faith); the repo frontmatter carries no license field. Skill Harbor never reviews the code, review it yourself before use. Discovered via skills.sh.
Version:
Install
Prerequisites: a machine learning project in Python (PyTorch and/or scikit-learn); an experiment-tracking backend (MLflow or Weights & Biases) if you want run logging Install "ML engineer — end-to-end machine learning guidance: PyTorch, scikit-learn, evaluation, MLOps" for me. It gives my agent @rightnow-ai's ML practitioner playbook: start from a strong simple baseline, evaluate with business-aligned metrics on held-out data, version datasets/code/hyperparameters/artifacts, build idempotent resumable training pipelines (PyTorch training loops, scikit-learn Pipelines with ColumnTransformers, GridSearchCV/Optuna tuning), track every run in MLflow or Weights & Biases, promote models through registry stages, and monitor production for drift with a rollback plan. Apache-2.0 licensed (per the discovery manifest; the repo frontmatter carries no license field). Repository: https://github.com/rightnow-ai/openfang/blob/main/crates/openfang-skills/bundled/ml-engineer/SKILL.md 1. Fetch the SKILL.md file (and any helper files) from the repository path into a temporary folder and summarize what it does in one or two sentences. 2. Safety check: review the SKILL.md and scripts for anything suspicious (unexpected network calls, shell commands, credential harvesting). This repo should contain zero secrets in code. Verify that holds here; STOP on any red flag and tell me. 3. Install it as a skill: copy SKILL.md and its helper files into the agent's skills directory, in a folder named "ml-engineer". 4. Verify with no network calls: frontmatter valid, files in place. 5. Report what was installed, where, and what I still need to do myself (e.g. set up my PyTorch/scikit-learn environment and experiment-tracking backend, then describe the ML problem to the agent). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.
Questions
How do I install a build?
Every product page includes a copy-paste install prompt. Paste it into your Muse and it sets the build up for you — no manual configuration.
Where does my money go?
Straight to the seller. Skill Harbor never processes payments: checkout happens on the seller’s own page, usually Stripe.
What does the ✓ next to a creator’s name mean?
It means we confirmed the identity of the person behind the listing. It says nothing about the code itself — always check a build before installing it.