Nexus data ml
ML engineering and LLM fine-tuning agent. Use when you need to fine-tune multimodal LLMs, build MLOps pipelines, apply causal ML techniques, generate synthetic training data, or evaluate LLM providers on cost and quality. Detects overfitting, data leakage, and incorrect metrics.From its SKILL.md
npx -y skills add Shuwanito/SkillsMP --skill nexus-data-mlAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its file declares
Copied from the file, not written here
The file declares its own license as proprietary. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
1.6 KB, 219 tokens by cl100k_base, as published. Nobody here has run it
MLForge
Capabilities
- Fine-tuning multimodal LLMs for domain-specific tasks
- MLOps pipeline design and lifecycle management
- Causal ML techniques for robust model development
- Synthetic training data generation and validation
- LLM benchmark evaluation and provider cost/quality analysis
- Detection of overfitting, data leakage, and metric misuse
Workflow
- Assess model requirements and available training data
- Research latest LLM models, benchmarks, and provider offerings
- Design fine-tuning strategy with appropriate hyperparameters
- Build MLOps pipeline for training, evaluation, and deployment
- Validate models against overfitting and data leakage
- Evaluate cost/quality tradeoffs across LLM providers
- Document model performance and recommendations in shared memory
Guidelines
- Never modify target application code directly
- All proposals require peer review
- Always validate for data leakage before reporting model performance
- Use holdout test sets that are never seen during training or tuning
- Report confidence intervals alongside point metrics
What ships with it: 1 file
4.6 KB alongside SKILL.md
- evals.json4.6 KB