Backend

GPTClaudeDeepSeek··▲ 1,252 copies·updated 2026-07-14

backend.prompt

Now implement Prompt 2 — Backend Layer only. Refine and improve the backend files you already created. Focus on making them production-ready and bug-free.

Build only the backend layer of a Developer Profile Analyzer.

Tech: Next.js 14 API routes, TypeScript, Gemini 2.0 Flash API, GitHub REST API.

Create these files:

1. src/lib/github/fetcher.ts

   - fetchUserProfile(username): calls /users/:username
   - fetchUserRepos(username): calls /users/:username/repos, filters out forks
   - fetchUserEvents(username): calls /users/:username/events/public
   - Handle 404, 403, network errors
   - In-memory cache with 24h TTL using Map

2. src/lib/analyzers/languageAnalyzer.ts

   - Input: repo list
   - Output: { name: string, percentage: number }[] sorted by usage

3. src/lib/analyzers/activityAnalyzer.ts

   - Input: events list
   - Output: { avgCommitsPerWeek: number, activeReposLast90Days: number, consistencyScore: number }

4. src/lib/analyzers/engineeringAnalyzer.ts

   - Input: repo list
   - Detect: hasTests, hasDockerfile, hasCICD, hasReadme, hasDeployment
   - Method: check repo file tree via /repos/:owner/:repo/contents
   - Output: EngineeringSignals interface

5. src/lib/scoring/scoreEngine.ts

   - Input: NormalizedProfile
   - Output: DeveloperScores { backend, frontend, devops, testing, consistency, projectDepth, overallScore }
   - All scores 0-100, computed deterministically

6. src/lib/ai/insightGenerator.ts

   - Input: NormalizedProfile + DeveloperScores
   - Call Gemini 2.0 Flash with structured prompt
   - Return: AIInsights { summary, strengths, weaknesses, recommendations, careerFit }

7. src/app/api/analyze/route.ts

   - POST endpoint, body: { username: string }
   - Orchestrates all above modules
   - Returns complete analysis as JSON

8. src/types/index.ts

   - All TypeScript interfaces

Include all imports. Make it production-ready. No placeholder code.

fill the variables

This prompt has 3 variables. Pro fills them into a ready-to-paste prompt for you — no manual find-and-replace.

{name: string, percentage: number}{summary, strengths, weaknesses, recommendations, careerFit}{username: string}

Unlock with Pro →

when to use it

Community prompt sourced from the open-source GitHub repo SuleymanToklu/LLM-Comparison-Benchmarking (no explicit license). A "Backend" style prompt — adapt the placeholders and specifics to your task. Imported as-is and not independently retested here, so check the output before relying on it.