Create a Tailored Resume for a Job Application

Your Task

When the user wants to create a tailored resume for a specific job:

1. •Get the job description - Ask the user to provide the job posting/description
2. •Analyze job requirements - Identify key skills, technologies, and experience mentioned
3. •Match with profile - Cross-reference job requirements with candidate's skills in content.typ
4. •Create tailored config.json - Select only relevant skills and rewrite experience bullets

Step 1: Get the Job Description

Ask the user: "Please paste the job description or provide a link to the job posting."

Step 2: Analyze Job Requirements

Read the job description and identify:

• •Required skills/technologies (e.g., PyTorch, NLP, Kubernetes, etc.)
• •Preferred skills (nice to have)
• •Experience level (e.g., "5+ years", "senior", "lead")
• •Key responsibilities (what they'll be working on)
• •Domain/industry (fintech, healthcare, ML infrastructure, etc.)

Step 3: Match with Candidate Profile

From content.typ, you have access to:

Skills available:

• •AI/ML: DSPy, PyTorch, Scikit-Learn, NVIDIA Triton, NLP, CV, DL
• •Backend: Python, FastAPI, NestJS, Litestar, RESTful APIs
• •Data: Pandas, MongoDB, PostgreSQL, Redis, RabbitMQ, SurrealDB
• •Frontend: Typescript, Next.js, React, Remix, TailwindCSS
• •Cloud/DevOps: Docker, AWS, Azure, CI/CD, Prometheus, Grafana, Nix
• •Languages: Romanian, English, German

Work experience (from content.typ work_experience):

• •AI Engineer at CogniSync (Sep 2025 - Present)
• •Co-Founder & CTO at Pythia (Jun 2022 - Sep 2025)
• •IT Administrator at Cert Sud (Sep 2020 - Sep 2022)
• •R&D Intern at Continental (Jun 2019 - Aug 2019)

Education:

• •M.Sc. Artificial Intelligence - University of Bucharest
• •B.Eng. Applied Electronics - Polytechnics University of Bucharest
• •Various Erasmus programs

Step 4: Build the config.json

A. Select Relevant Skills

For each skill category in skills_data, include only skills that:

1. •Are mentioned in the job description, OR
2. •Are highly relevant to the role even if not explicitly mentioned

Example for an ML Engineer role:

"selected_skills": {
  "AI/ML": ["DSPy", "PyTorch", "NLP", "DL"],
  "Backend": ["Python", "FastAPI", "RESTful APIs"],
  "Data": ["Pandas", "MongoDB", "RabbitMQ"],
  "Languages": ["English", "Romanian"]
}

B. Rewrite Work Experience Bullets

For each job, select bullet points that:

1. •Directly relate to job requirements
2. •Use keywords from the job posting
3. •Quantify achievements (percentages, numbers, time saved)

DO NOT rephrase existing bullets unnecessarily. Only select the most relevant ones from content.typ work_experience. If needed, you can slightly adjust wording to better match job keywords.

Example:

"work_experience": [
  {
    "title": "Co-Founder & CTO",
    "company": "Pythia Sociodynamical Technologies",
    "location": "Bucharest, Romania",
    "dates": {"start": "Jun 2022", "end": "Sep 2025"},
    "bullets": [
      "Architected NVIDIA Triton-based ML inference pipeline processing 1,000+ news articles/sec with <100ms latency, reducing infrastructure costs by 35%",
      "Engineered LLM & DSPy powered data extraction with intelligent content hashing, increasing accuracy by >40%"
    ]
  }
]

C. Write Professional Summary

Write 3 lines that:

• •Directly address the job requirements
• •Highlight matching skills and experience
• •Use keywords from the job description
• •Avoid first-person pronouns (I, me, my)
• •Focus on how the candidate's background solves the employer's problems

Step 5: Create Files

mkdir -p applications/CompanyName-JobTitle
cp templates/application_template.typ applications/CompanyName-JobTitle/resume.typ

Create config.json with:

• •position: The job title
• •professional_summary: 3-line pitch matching the job
• •include_sections: Sections to include (typically ["professional_summary", "skills", "work_experience", "education"])
• •selected_skills: Only skills matching the job
• •work_experience: Only relevant experience with tailored bullets

Key Guidelines

1. •DO NOT include irrelevant skills - If the job doesn't mention React, don't include Frontend skills
2. •DO include niche relevant skills - If the job needs DSPy or NVIDIA Triton, highlight those
3. •Keep bullets concise - 3-5 bullets per job entry
4. •Use ATS keywords - Mirror language from the job posting
5. •Quantify achievements - Keep metrics (35% cost reduction, 99.9% uptime, etc.)
6. •Order by relevance - Most relevant job entries first

Example config.json

{
  "position": "Senior Machine Learning Engineer",
  "professional_summary": "Machine Learning Engineer with 3+ years designing production ML systems. Expert in LLM-powered applications, Triton inference optimization, and scalable data pipelines. Reduced ML infrastructure costs by 35% while processing 1M+ daily transactions.",
  "include_sections": ["professional_summary", "skills", "work_experience", "education"],
  "selected_skills": {
    "AI/ML": ["DSPy", "PyTorch", "NVIDIA Triton", "NLP", "LLMs"],
    "Backend": ["Python", "FastAPI", "RESTful APIs"],
    "Data": ["Pandas", "MongoDB", "RabbitMQ"],
    "Cloud/DevOps": ["Docker", "AWS", "Prometheus", "Grafana"],
    "Languages": ["English", "Romanian"]
  },
  "work_experience": [
    {
      "title": "Co-Founder & CTO",
      "company": "Pythia Sociodynamical Technologies",
      "location": "Bucharest, Romania",
      "dates": {"start": "Jun 2022", "end": "Sep 2025"},
      "bullets": [
        "Architected NVIDIA Triton-based ML inference pipeline processing 1,000+ items/sec with <100ms latency, reducing infrastructure costs by 35%",
        "Engineered LLM-powered data extraction using DSPy, improving accuracy by 40% while reducing inference costs",
        "Deployed comprehensive observability stack with Prometheus and Grafana, maintaining 99.9% system uptime"
      ]
    },
    {
      "title": "AI Engineer",
      "company": "CogniSync",
      "location": "Bucharest, Romania",
      "dates": {"start": "Sep 2025", "end": "Present"},
      "bullets": [
        "Engineered AI system for processing and understanding large legal documents using modern LLM techniques"
      ]
    }
  ]
}

Compiling the Resume

make app applications/CompanyName-JobTitle
make compile-app applications/CompanyName-JobTitle
make watch-app applications/CompanyName-JobTitle