test assignment reviewer

The problem

Reviewing technical test assignments is a time-consuming bottleneck for engineering teams. Manually cross-referencing a candidate's repository against a multi-page task description often leads to inconsistent scoring and missed red flags.

What it does

• Parses task descriptions from GitHub, PDFs, or Notion to extract mandatory features and constraints.
• Audits candidate implementation code for stack compliance, architecture patterns, and code quality.
• Generates a structured verdict (HIRE/NO HIRE) with a 60-point scoring matrix across six key signals.
• Produces a line-by-line gap analysis comparing specific requirements to actual implementation.
• Drafts 5-7 targeted interview questions based on identified deviations or suspicious code choices.

Frameworks & tools

Works with GitHub repositories, deployed applications, and document formats including PDF, Notion, and Google Docs. Uses Bash and WebFetch for deep repository inspection.

Why this beats prompting it yourself

This skill enforces a strict, evidence-based evaluation framework that prevents hallucinated praise or vague feedback. It automates the tedious task of requirement mapping, ensuring no mandatory feature is overlooked, and provides export-ready artifacts for your ATS or team wiki.

Use cases

• Reviewing full-stack take-home assignments for mid-to-senior level roles.
• Screening initial repository submissions before scheduling technical interviews.
• Generating standardized hiring reports for engineering managers and stakeholders.
• Audit candidate decision-making by identifying intentional deviations from the task prompt.

Known limitations

Evaluation of deployed apps without source code is limited to UX and feature completeness. ZIP archives must be described or pasted inline as direct file uploads are not supported.