Name: senior-ml-engineer
Rating: 68 (1 reviews)
Author: alirezarezvani

senior-ml-engineer

ML engineering skill for productionizing models, building MLOps pipelines, and integrating LLMs. Covers model deployment, feature stores, drift monitoring, RAG systems, and cost optimization. Use when the user asks about deploying ML models to production, setting up MLOps infrastructure (MLflow, Kubeflow, Kubernetes, Docker), monitoring model performance or drift, building RAG pipelines, or integrating LLM APIs with retry logic and cost controls. Focused on production and operational concerns rather than model research or initial training.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

The canonical home for this skill is senior-ml-engineer in alirezarezvani/claude-skills

Quality

Content

71%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is well-organized, actionable, and uses explicit validation checkpoints across its workflows. Its main weakness is progressive disclosure: the SKILL.md points to reference and script files that are not present in the bundle, leaving navigation broken.

Suggestions

Ship the referenced bundle files (references/mlops_production_patterns.md, references/llm_integration_guide.md, references/rag_system_architecture.md, and the three scripts/*.py) so the one-level-deep navigation actually resolves.

Add explicit error-recovery loops to the workflows (e.g., "if canary metrics fail, roll back and re-export") so validation steps close the loop rather than only stating pass criteria.

Complete the Feast code example by adding the `from feast import ValueType` and `from datetime import timedelta` imports so it is copy-paste runnable.

Dimension	Reasoning	Score
Conciseness	The body is reference-style and mostly lean — numbered workflows, tables, and code blocks earn their place — but the opening line restates the description and the Table of Contents adds tokens that could be trimmed, keeping it just short of fully lean.	4 / 5
Actionability	Provides executable artifacts — a Dockerfile, tenacity retry code, scipy drift detection, and concrete CLI commands with flags — but the Feast FeatureView snippet omits the `ValueType`/`timedelta` imports, a minor gap preventing copy-paste completeness.	4 / 5
Workflow Clarity	Each of the five workflows is a clear numbered sequence ending in an explicit "Validation: ..." checkpoint, and the canary-and-monitor deployment acts as a feedback loop; however, the workflows state pass/fail criteria without explicit error-recovery loops ("if validation fails, return to step N").	4 / 5
Progressive Disclosure	The overview is well structured with clearly signaled, one-level-deep references to `references/.md` and `scripts/.py`, but none of those bundle files actually exist in `references/`, `scripts/`, or `assets/`, so every reference is dangling and navigation to the detail material is broken.	3 / 5
	Total	15 / 20 Passed

Description

96%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, specific description that clearly states capabilities, natural trigger terms, and when to use the skill, with a useful scope boundary. The only weakness is some overlap with adjacent prompt-engineering and data-engineering skills.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "productionizing models", "building MLOps pipelines", "model deployment, feature stores, drift monitoring, RAG systems, cost optimization", "retry logic and cost controls" — giving comprehensive coverage rather than just naming the domain.	5 / 5
Completeness	Explicitly answers both "what" (productionizing models, MLOps, LLM integration) and "when" ("Use when the user asks about deploying ML models...") with concrete trigger phrases, plus a boundary clause ("Focused on production and operational concerns rather than model research or initial training").	5 / 5
Trigger Term Quality	Natural user phrases and concrete tool names are well covered: "deploying ML models to production", "MLOps infrastructure (MLflow, Kubeflow, Kubernetes, Docker)", "monitoring model performance or drift", "building RAG pipelines", "integrating LLM APIs", matching the comprehensive-coverage anchor.	5 / 5
Distinctiveness Conflict Risk	The production-MLOps niche is clear and the boundary clause reduces overlap, but LLM integration, RAG, and cost-optimization territory overlaps with prompt-engineering / data-engineering skills, leaving minor conflict risk with closely related skills.	4 / 5
	Total	19 / 20 Passed

Validation

87%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
referenced_paths_exist	Referenced path issues: 6 missing	Warning

	Total	14 / 16 Passed

Repository: alirezarezvani/claude-skills
Path: .gemini/skills/senior-ml-engineer/SKILL.md
Commit: 19392f7

Reviewed: ~10 hours ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.