Career Copilot — Sistema Operacional de Carreira para IA/ML
DhruvkrSharma/career-copilot-flat
Operating system de carreira para engenheiros de IA/ML com análise profunda e otimização
Ver no GitHubSkill abrangente que otimiza sua candidatura em IA/ML através de análise de ATS, otimização de currículo com proteção de pontuação, avaliação de GitHub e LinkedIn, análise de projetos, preparação para entrevistas e scoring de pronto para o mercado. Cobre tendências de 2026 como IA agêntica, LLMOps e MCP, com conhecimento atualizado de modelos, frameworks e mercado de trabalho.
Instalação
npx convez-ai DhruvkrSharma/career-copilot-flat1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.
Métricas
Casos de uso
- Otimizar currículo para ATS sem perder conteúdo original
- Avaliar perfil completo (GitHub, LinkedIn, projetos) para oportunidades em IA
- Preparar para entrevistas técnicas em IA/ML com tópicos 2026
- Encontrar gaps de keywords em candidatura e priorizar preenchimento
- Posicionar-se conforme tendências de mercado em engenharia de IA
Veredito

Ferramenta especializada e bem estruturada para profissionais de IA/ML que buscam otimizar sua candidatura de forma sistemática. O foco em proteção de ATS score e conhecimento atualizado (junho 2026) a torna relevante, embora dependa de prompts bem formulados e dados atualizados do usuário para máxima efetividade.
README
A tradução para o português deste README ainda está na fila.
🚀 Career Copilot — AI/ML Career Operating System
Last Updated: June 2026
A comprehensive AI Skill for AI/ML career optimization. ATS analysis with keyword preservation, resume optimization, keyword research, GitHub review, LinkedIn review, project evaluation, interview preparation, market fit analysis, and career readiness scoring — all updated to the June 2026 AI/ML landscape.
📋 What This Does
Career Copilot is a complete career operating system for AI/ML engineers. Instead of evaluating just your resume, it evaluates your entire candidate package:
- Resume — ATS compatibility, keyword density, multi-placement scoring, keyword preservation
- GitHub — Profile presentation, code quality, documentation
- LinkedIn — Recruiter visibility, professional brand, network
- Projects — Technical depth, deployment, evaluation pipelines, observability
- Interview Readiness — STAR answers, agentic AI design, system design, evaluation skills
- Market Fit — Alignment with 2026 hiring trends (agentic AI, LLMOps, multi-modal, MCP)
✨ Key Features (June 2026)
🛡️ ATS Score Protection System
The career optimize command guarantees your ATS score never decreases during optimization:
- Baseline scoring — Records your current ATS score before any changes
- Keyword preservation — Catalogs every existing keyword and locks them in
- Gap filling — Adds missing keywords from the job description and role bank
- Multi-placement — Ensures keywords appear in Skills AND contextual bullets
- Post-rewrite verification — Re-scores and blocks output if ATS score dropped
- Score gate loop — If score dropped, automatically revises until it's ≥ baseline
🔥 2026-Optimized Knowledge Base
- Latest models: GPT-5.5, Claude Opus 4.8, Gemini 3.1 Pro, Llama 4, Grok-3
- Latest frameworks: LangGraph, CrewAI, OpenAI Agents SDK, Google ADK, SGLang
- Emerging paradigms: Agentic RAG, GraphRAG, MCP/A2A protocols, Reasoning Models
- New roles covered: AI Infrastructure Engineer, AI Product Engineer, AI Safety Engineer
- Updated salaries & market data for India and US (June 2026)
🎯 Commands
| Command | Description |
|---|---|
career diagnose | Quick 5-minute triage with traffic-light scoring |
career keywords | Deep keyword gap analysis with multi-placement matrix and priority ranking |
career optimize | Resume optimization with ATS score protection and keyword preservation |
career github | GitHub profile review and scoring |
career linkedin | LinkedIn profile review and scoring |
career projects | Project portfolio evaluation (now includes agentic AI archetypes) |
career interview | Technical + behavioral interview simulation (2026 topics) |
career full-review | Complete career package evaluation → Career Score /100 |
📁 Folder Structure
career-copilot/
├── SKILL.md # Entry point — commands, persona, file routing
│
├── knowledge/
│ ├── ai_ml_keywords.md # Keywords by role — updated June 2026 (9 roles)
│ ├── project_patterns.md # Project signals + 2026 S-tier archetypes
│ ├── interview_knowledge.md # Interview topics — agentic AI, LLMOps, AI safety
│ ├── resume_best_practices.md # Resume rules, ATS optimization, keyword preservation
│ ├── market_intelligence.md # 2026 hiring trends, salaries, hot skills
│ └── my_profile.md # Personal profile template (fill in your details)
│
├── rubrics/
│ ├── ai_resume_rubric.md # 100-point resume scoring (ATS boosted to 20pts)
│ ├── github_rubric.md # 100-point GitHub scoring (7 categories)
│ ├── interview_rubric.md # 100-point interview scoring (6 categories)
│ ├── project_evaluation_rubric.md # 100-point project scoring (9 categories)
│ └── linkedin_rubric.md # 100-point LinkedIn scoring (7 categories)
│
├── prompts/
│ ├── career_full_review.md # Orchestration prompt (flagship command)
│ ├── career_diagnose.md # Quick triage diagnostic
│ ├── ats_review.md # ATS analysis — 100pt scoring + re-verification mode
│ ├── keyword_analysis.md # Deep keyword gap analysis with placement matrix
│ ├── resume_rewrite.md # Resume rewriting with keyword preservation loop
│ ├── github_review.md # GitHub profile evaluation
│ ├── linkedin_review.md # LinkedIn profile evaluation
│ ├── project_review.md # Project portfolio evaluation
│ └── interview.md # Interview simulation
│
└── examples/
├── good_resume_example.md # Weak vs strong bullets + ATS preservation examples
├── strong_project_example.md # S-tier project benchmark
├── interview_answer_example.md # STAR method demonstrations
├── weak_github_example.md # Bad vs good GitHub profiles
├── linkedin_example.md # Weak vs strong LinkedIn profiles
└── system_design_answer.md # ML system design answers
🔧 Installation
For Claude Code / Gemini (Skills)
# Claude Code
mkdir -p ~/.claude/skills/
cp -r career-copilot/ ~/.claude/skills/career-copilot/
# Gemini (Antigravity)
# Place in your configured skills directory
Personalization
Edit knowledge/my_profile.md with your actual details to enable personalized scoring.
📊 Full Review Output
Running career full-review produces:
═══════════════════════════════════════════
CAREER READINESS REPORT
═══════════════════════════════════════════
Career Score: __/100 [Elite|Strong|Good|Average|Weak]
┌─────────────────────────────────────────┐
│ ATS Score ██████████░░ __/100 │
│ Project Score ████████░░░░ __/100 │
│ GitHub Score ██████░░░░░░ __/100 │
│ Interview Score ████████████ __/100 │
│ Market Fit Score ██████████░░ __/100 │
│ LinkedIn Score ████████░░░░ __/100 │
└─────────────────────────────────────────┘
🏆 Top Strengths
⚠️ Critical Gaps
📋 30-Day Action Plan
📋 90-Day Roadmap
🎯 Target Readiness
🎓 Target Roles
Optimized for:
- AI Engineer
- Machine Learning Engineer
- Computer Vision Engineer
- Data Scientist
- MLOps / LLMOps Engineer
- Software Engineer
- GenAI / Agentic AI Engineer
- AI Infrastructure Engineer (new)
- AI Product Engineer (new)
📈 What Makes This Different
Unlike simple resume review tools, Career Copilot:
- Evaluates the full package — Resume + GitHub + LinkedIn + Projects + Interview readiness
- Protects your ATS score — Keyword preservation system prevents score drops during optimization
- Uses structured rubrics — 100-point scoring with transparent criteria
- Market-aware — Updated to June 2026 trends (agentic AI, MCP, LLMOps, multi-modal)
- Deep keyword analysis — Multi-placement matrix, JD mirroring, density scoring, priority ranking
- Personalized — Fill in your profile for tailored recommendations
- Actionable — Every weakness comes with a concrete fix and timeline
📝 Changelog
June 2026 — Major Update
ATS Score Protection (Bug Fix)
- Fixed critical bug where resume optimization was reducing ATS scores instead of increasing them
- Root cause: Rewrite process was replacing keywords with synonyms (e.g., "TensorFlow" → "deep learning framework")
- Added 5-step keyword preservation pipeline: Extract → Preserve → Add → Verify → Gate
- Added post-rewrite ATS re-verification with PASS/FAIL verdict
- Added
career optimizemandatory execution order (Baseline → Gap Analysis → Rewrite → Verify → Score Gate)
ATS Scoring Overhaul (Original Fix)
- ATS weight in resume rubric increased from 15 → 20 points
- ATS review originally added keyword density, multi-placement, and JD phrase matching scoring
Structural Architecture Fixes (Latest Update)
- Evidence-Based ATS Model: Replaced folk-theory keyword density targets with a parseability-focused, recruiter-search model. Focus is now on real parsing risks (formatting, fonts, unselectable text) and keyword presence rather than density.
- Rubric Calibration: Added clear, reproducible calibration anchors across all 5 rubrics so scores are consistent and not falsely precise. Included an honesty check to score metrics as-stated but flag implausible claims.
- Data Acquisition Preconditions: Wired in explicit data-acquisition instructions for GitHub and LinkedIn in the orchestration prompt. The model will now ask for the profile content rather than stalling or hallucinating.
- Centralized Rules: Consolidated keyword preservation and rewrite rules into
resume_best_practices.mdas the canonical source to prevent drift across multiple prompt files. - Data Decay Controls: Added explicit "verify live before quoting" staleness warnings to volatile data files (
market_intelligence.md,ai_ml_keywords.md) regarding salaries, model names, and market trends. - Contradiction Fix: Resolved contradictory guidance around System Design rounds (now scaled by level rather than gated).
Knowledge Base — Updated to June 2026
ai_ml_keywords.md— Latest models (GPT-5.5, Claude Opus 4.8, Gemini 3.x, Llama 4), new frameworks (LangGraph, OpenAI Agents SDK, Google ADK), new paradigms (MCP, A2A, Agentic RAG), 2 new rolesmarket_intelligence.md— 2026 salary data, agentic AI as #1 skill, new roles, updated company lists, interview trend shiftsinterview_knowledge.md— Agentic AI interview section, LLMOps, AI Safety, reasoning models, 2026-specific behavioral themesproject_patterns.md— 5 S-tier project archetypes, observability requirements, overused projects to avoid, updated tier classificationsresume_best_practices.md— Separate section priority for GenAI vs ML roles, 2026 strong examples (agentic systems, model routing, fine-tuning), cost metrics, evaluation maturity, observability
📝 License
MIT