resume generator cmdline utility

.gitignore

@@ -11,3 +11,5 @@ generated_cv*
 .vscode
 chrome_profile
 answers.json
+resume.yaml
+resume.yaml_validation_report.txt

README.md

@@ -452,6 +452,27 @@ Each section has specific fields to fill out:
     willing_to_undergo_drug_tests: "No"
     willing_to_undergo_background_checks: "Yes"
   ```
+### 4. Generating plain_text_resume.yaml from a Text Resume
+
+To simplify the process of creating your `plain_text_resume.yaml` file, you can use the provided script to generate it from a text-based resume. Follow these steps:
+
+1. Prepare your resume in a plain text format (.txt file).
+
+2. Place your text resume in the `data_folder` directory.
+
+3. Run the following command:
+
+   ```bash
+   python generate_resume_yaml.py --input data_folder/your_resume.txt --output data_folder/plain_text_resume.yaml
+   ```
+
+   Replace `your_resume.txt` with the actual name of your text resume file.
+
+4. The script will generate a `plain_text_resume.yaml` file in the `data_folder` directory.
+
+5. Review the generated YAML file and make any necessary adjustments to ensure all information is correct and complete.
+
+This automated process helps in creating a structured YAML file from your existing resume, saving time and reducing the chance of errors in manual data entry.
 
 ### PLUS. data_folder_example
 

assets/resume_liam_murphy.txt

@@ -0,0 +1,55 @@
+Liam Murphy
+Galway, Ireland
+Email: liam.murphy@gmail.com | LinkedIn: liam-murphy
+GitHub: liam-murphy | Phone: +353 871234567
+
+Education
+Bachelor's Degree in Computer Science
+National University of Ireland, Galway (GPA: 4/4)
+Graduation Year: 2020
+
+Experience
+Co-Founder & Software Engineer
+CryptoWave Solutions (03/2021 - Present)
+Location: Ireland | Industry: Blockchain Technology
+
+Co-founded and led a startup specializing in app and software development with a focus on blockchain technology
+Provided blockchain consultations for 10+ companies, enhancing their software capabilities with secure, decentralized solutions
+Developed blockchain applications, integrated cutting-edge technology to meet client needs and drive industry innovation
+Research Intern
+National University of Ireland, Galway (11/2022 - 03/2023)
+Location: Galway, Ireland | Industry: IoT Security Research
+
+Conducted in-depth research on IoT security, focusing on binary instrumentation and runtime monitoring
+Performed in-depth study of the MQTT protocol and Falco
+Developed multiple software components including MQTT packet analysis library, Falco adapter, and RML monitor in Prolog
+Authored thesis "Binary Instrumentation for Runtime Monitoring of Internet of Things Systems Using Falco"
+Software Engineer
+University Hospital Galway (05/2022 - 11/2022)
+Location: Galway, Ireland | Industry: Healthcare IT
+
+Integrated and enforced robust security protocols
+Developed and maintained a critical software tool for password validation used by over 1,600 employees
+Played an integral role in the hospital's cybersecurity team
+Projects
+JobBot
+AI-driven tool to automate and personalize job applications on LinkedIn, gained over 3000 stars on GitHub, improving efficiency and reducing application time
+Link: JobBot
+
+mqtt-packet-parser
+Developed a Node.js module for parsing MQTT packets, improved parsing efficiency by 40%
+Link: mqtt-packet-parser
+
+Achievements
+Winner of an Irish public competition - Won first place in a public competition with a perfect score of 70/70, securing a Software Developer position at University Hospital Galway
+Galway Merit Scholarship - Awarded annually from 2018 to 2020 in recognition of academic excellence and contribution
+GitHub Recognition - Gained over 3000 stars on GitHub with JobBot project
+Certifications
+C1
+
+Languages
+English - Native
+Spanish - Professional
+Interests
+Full-Stack Development, Software Architecture, IoT system design and development, Artificial Intelligence, Cloud Technologies
+

assets/resume_schema.yaml

@@ -0,0 +1,132 @@
+# YAML Schema for plain_text_resume.yaml
+
+personal_information:
+  type: object
+  properties:
+    name: {type: string}
+    surname: {type: string}
+    date_of_birth: {type: string, format: date}
+    country: {type: string}
+    city: {type: string}
+    address: {type: string}
+    phone_prefix: {type: string, format: phone_prefix}
+    phone: {type: string, format: phone}
+    email: {type: string, format: email}
+    github: {type: string, format: uri}
+    linkedin: {type: string, format: uri}
+  required: [name, surname, date_of_birth, country, city, address, phone_prefix, phone, email]
+
+education_details:
+  type: array
+  items:
+    type: object
+    properties:
+      degree: {type: string}
+      university: {type: string}
+      gpa: {type: string}
+      graduation_year: {type: string}
+      field_of_study: {type: string}
+      exam:
+        type: object
+        additionalProperties: {type: string}
+    required: [degree, university, gpa, graduation_year, field_of_study]
+
+experience_details:
+  type: array
+  items:
+    type: object
+    properties:
+      position: {type: string}
+      company: {type: string}
+      employment_period: {type: string}
+      location: {type: string}
+      industry: {type: string}
+      key_responsibilities:
+        type: object
+        additionalProperties: {type: string}
+      skills_acquired:
+        type: array
+        items: {type: string}
+    required: [position, company, employment_period, location, industry, key_responsibilities, skills_acquired]
+
+projects:
+  type: array
+  items:
+    type: object
+    properties:
+      name: {type: string}
+      description: {type: string}
+      link: {type: string, format: uri}
+    required: [name, description]
+
+achievements:
+  type: array
+  items:
+    type: object
+    properties:
+      name: {type: string}
+      description: {type: string}
+    required: [name, description]
+
+certifications:
+  type: array
+  items: {type: string}
+
+languages:
+  type: array
+  items:
+    type: object
+    properties:
+      language: {type: string}
+      proficiency: {type: string, enum: [Native, Fluent, Intermediate, Beginner]}
+    required: [language, proficiency]
+
+interests:
+  type: array
+  items: {type: string}
+
+availability:
+  type: object
+  properties:
+    notice_period: {type: string}
+  required: [notice_period]
+
+salary_expectations:
+  type: object
+  properties:
+    salary_range_usd: {type: string}
+  required: [salary_range_usd]
+
+self_identification:
+  type: object
+  properties:
+    gender: {type: string}
+    pronouns: {type: string}
+    veteran: {type: string, enum: [Yes, No]}
+    disability: {type: string, enum: [Yes, No]}
+    ethnicity: {type: string}
+  required: [gender, pronouns, veteran, disability, ethnicity]
+
+legal_authorization:
+  type: object
+  properties:
+    eu_work_authorization: {type: string, enum: [Yes, No]}
+    us_work_authorization: {type: string, enum: [Yes, No]}
+    requires_us_visa: {type: string, enum: [Yes, No]}
+    requires_us_sponsorship: {type: string, enum: [Yes, No]}
+    requires_eu_visa: {type: string, enum: [Yes, No]}
+    legally_allowed_to_work_in_eu: {type: string, enum: [Yes, No]}
+    legally_allowed_to_work_in_us: {type: string, enum: [Yes, No]}
+    requires_eu_sponsorship: {type: string, enum: [Yes, No]}
+  required: [eu_work_authorization, us_work_authorization, requires_us_visa, requires_us_sponsorship, requires_eu_visa, legally_allowed_to_work_in_eu, legally_allowed_to_work_in_us, requires_eu_sponsorship]
+
+work_preferences:
+  type: object
+  properties:
+    remote_work: {type: string, enum: [Yes, No]}
+    in_person_work: {type: string, enum: [Yes, No]}
+    open_to_relocation: {type: string, enum: [Yes, No]}
+    willing_to_complete_assessments: {type: string, enum: [Yes, No]}
+    willing_to_undergo_drug_tests: {type: string, enum: [Yes, No]}
+    willing_to_undergo_background_checks: {type: string, enum: [Yes, No]}
+  required: [remote_work, in_person_work, open_to_relocation, willing_to_complete_assessments, willing_to_undergo_drug_tests, willing_to_undergo_background_checks]

resume_yaml_generator.py

@@ -0,0 +1,164 @@
+import argparse
+import yaml
+from openai import OpenAI
+import os
+from typing import Dict, Any
+import tiktoken
+import re
+from jsonschema import validate, ValidationError
+
+def load_yaml(file_path: str) -> Dict[str, Any]:
+    with open(file_path, 'r') as file:
+        return yaml.safe_load(file)
+
+def load_resume_text(file_path: str) -> str:
+    with open(file_path, 'r') as file:
+        return file.read()
+
+def get_api_key() -> str:
+    secrets_path = os.path.join('data_folder', 'secrets.yaml')
+    if not os.path.exists(secrets_path):
+        raise FileNotFoundError(f"Secrets file not found at {secrets_path}")
+    
+    secrets = load_yaml(secrets_path)
+    api_key = secrets.get('openai_api_key')
+    if not api_key:
+        raise ValueError("OpenAI API key not found in secrets.yaml")
+    
+    return api_key
+
+def num_tokens_from_string(string: str, model: str) -> int:
+    encoding = tiktoken.encoding_for_model(model)
+    return len(encoding.encode(string))
+
+def generate_yaml_from_resume(resume_text: str, schema: Dict[str, Any], api_key: str) -> str:
+    client = OpenAI(api_key=api_key)
+
+    prompt = f"""
+    I'm sending you the content of a text-based resume. Your task is to interpret this content and generate a YAML file that conforms to the following schema structure.
+    The generated YAML should include all required fields and follow the structure defined in the schema.
+
+    Pay special attention to the property attributes in the schema. These indicate the expected type and format for each field:
+    - 'type': Specifies the data type (e.g., string, object, array)
+    - 'format': Indicates a specific format for certain fields:
+    - 'date' format should be a valid date (e.g., YYYY-MM-DD)
+    - 'phone_prefix' format should be a valid country code with a '+' prefix (e.g., +1 for US)
+    - 'phone' format should be a valid phone number
+    - 'email' format should be a valid email address
+    - 'uri' format should be a valid URL
+    - 'enum': Provides a list of allowed values for a field
+
+    Important instructions:
+    1. Ensure that the YAML structure matches exactly with the provided schema. Use a dictionary structure that mirrors the schema.
+    2. For all sections, if information is not explicitly provided in the resume, make a best guess based on the context of the resume. This is CRUCIAL for the following fields:
+    - languages: Infer from the resume content or make an educated guess. Use the 'enum' values for proficiency.
+    - interests: Deduce from the overall resume or related experiences.
+    - availability (notice_period): Provide a reasonable estimate (e.g., "2 weeks" or "1 month").
+    - salary_expectations (salary_range_usd): Estimate based on experience level and industry standards.
+    - self_identification: Make reasonable assumptions based on the resume context. Use 'enum' values where provided.
+    - legal_authorization: Provide plausible values based on the resume information. Use 'Yes' or 'No' as per the 'enum' values.
+    - work_preferences: Infer from job history, skills, and overall resume tone. Use 'Yes' or 'No' as per the 'enum' values.
+    3. For the fields mentioned in point 2, always provide a value. Do not leave them blank or omit them.
+    4. For the 'key_responsibilities' field in 'experience_details', format the responsibilities as follows:
+    responsibility_1: "Description of first responsibility"
+    responsibility_2: "Description of second responsibility"
+    responsibility_3: "Description of third responsibility"
+    responsibility_4: "Description of fourth responsibility"
+    Continue this pattern for all responsibilities listed.
+    5. In the 'experience_details' section, ensure that 'position' comes before 'company' in each entry.
+    6. For the 'skills_acquired' field in 'experience_details', infer relevant skills based on the job responsibilities and industry. Do not leave this field empty.
+    7. Make reasonable inferences for any missing dates, such as date_of_birth or employment dates, ensuring they follow the 'date' format.
+    8. For array types (e.g., education_details, experience_details), ensure to include all required fields for each item as specified in the schema.
+
+    Resume Text Content:
+    {resume_text}
+
+    YAML Schema:
+    {yaml.dump(schema, default_flow_style=False)}
+
+    Generate the YAML content that matches this schema based on the resume content provided, ensuring all format hints are followed and making educated guesses where necessary. Be sure to include best guesses for ALL fields, even if not explicitly mentioned in the resume.
+    Enclose your response in <resume_yaml> tags. Only include the YAML content within these tags, without any additional text or code block markers.
+    """
+    
+    model = "gpt-3.5-turbo-16k"  # This model has a 16k token limit
+    tokens = num_tokens_from_string(prompt, model)
+    max_tokens = min(16385 - tokens, 4000)  # Ensure we don't exceed model's limit
+
+    if tokens > 16385:
+        print(f"Warning: The input exceeds the model's context length. Tokens: {tokens}")
+
+    response = client.chat.completions.create(
+        model=model,
+        messages=[
+            {"role": "system", "content": "You are a helpful assistant that generates structured YAML content from resume files, paying close attention to format requirements and schema structure."},
+            {"role": "user", "content": prompt}
+        ],
+        max_tokens=max_tokens,
+        n=1,
+        stop=None,
+        temperature=0.5,
+    )
+
+    yaml_content = response.choices[0].message.content.strip()
+    
+    # Extract YAML content from between the tags
+    match = re.search(r'<resume_yaml>(.*?)</resume_yaml>', yaml_content, re.DOTALL)
+    if match:
+        return match.group(1).strip()
+    else:
+        raise ValueError("YAML content not found in the expected format")
+                                                                                                                                                                                                                                                                                                                                                                                        
+def save_yaml(data: str, output_file: str):
+    with open(output_file, 'w') as file:
+        file.write(data)
+
+def validate_yaml(yaml_content: str, schema: Dict[str, Any]) -> Dict[str, Any]:
+    try:
+        yaml_dict = yaml.safe_load(yaml_content)
+        validate(instance=yaml_dict, schema=schema)
+        return {"valid": True, "errors": None}
+    except ValidationError as e:
+        return {"valid": False, "errors": str(e)}
+
+def generate_report(validation_result: Dict[str, Any], output_file: str):
+    report = f"Validation Report for {output_file}\n"
+    report += "=" * 40 + "\n"
+    if validation_result["valid"]:
+        report += "YAML is valid and conforms to the schema.\n"
+    else:
+        report += "YAML is not valid. Errors:\n"
+        report += validation_result["errors"] + "\n"
+    
+    print(report)
+    with open(f"{output_file}_validation_report.txt", 'w') as file:
+        file.write(report)
+
+def main():
+    parser = argparse.ArgumentParser(description="Generate a resume YAML file from a text resume using OpenAI API")
+    parser.add_argument("resume_file", help="Path to the input text resume file")
+    parser.add_argument("schema_file", help="Path to the YAML schema file")
+    parser.add_argument("output_file", help="Path to the output YAML file")
+    args = parser.parse_args()
+
+    try:
+        api_key = get_api_key()
+        schema = load_yaml(args.schema_file)
+        resume_text = load_resume_text(args.resume_file)
+
+        generated_yaml = generate_yaml_from_resume(resume_text, schema, api_key)
+        save_yaml(generated_yaml, args.output_file)
+
+        print(f"Resume YAML generated and saved to {args.output_file}")
+
+        validation_result = validate_yaml(generated_yaml, schema)
+        generate_report(validation_result, args.output_file)
+
+    except FileNotFoundError as e:
+        print(f"Error: {e}")
+    except ValueError as e:
+        print(f"Error: {e}")
+    except Exception as e:
+        print(f"An unexpected error occurred: {e}")
+
+if __name__ == "__main__":
+    main()