project-proposal-2025

JobFitAI

Abstract

If you’ve ever struggled to get your resume via automated recruiting processes or wanted expert advice on how to improve it, JobFitAI is here to help. Imagine uploading your resume and instantly seeing how it stacks up against job descriptions in your field. With JobFitAI, you’ll not only pass automated screens but also stand out to human reviewers.

This web-based tool uses Artificial Intelligence to analyse your resume and provide you with specific ideas to match what companies and applicant tracking systems (ATS) are searching for. JobFitAI makes it simple to improve your resume by removing formatting errors, adding missing keywords, and highlighting achievements. JobFitAI is designed to manage thousands of users safely and scale during peak job-hunting seasons, ensuring your personal information remains secure while assisting you in landing more interviews.

Author

Name: Jival Kapoor

Student number: 48245764

Functionality

JobFitAI’s complete system will include:

Resume Analysis: The AI-driven tool evaluates resumes for ATS compatibility, keyword alignment with job descriptions, formatting errors and readability. it checks for missing sections (for eg. Summary or Skills), inconsistent bullet points, and font sizes that might confuse ATS parsers.
Providing Actionable Feedback: Provides proritized suggestions for improving ATS scores to better align with the job description provided by the user (for example “Include ‘python’ as a skill to match 80% of data scientist job postings”)
Secure Storage: Resumes are encrypted during upload and storage ensuring sensitive details like phone numbers and addresses remain private.
Multi-Format Support: The tool accepts PDF, DOCX, and plain text files, accommodating users who draft resumes in different formats.
Future compatibility with LinkedIn and job boards for real-time job matching.

Scope (MVP)

The Minimum Viable Product will deliver:

Web Interface for Resume Upload & Analysis
- Users access the tool via a browser to upload resumes in PDF or DOCX format on a dashboard.
- The AI scans the document and provides an instant ATS compatibility score (0–100), flagging critical issues like missing contact information or unreadable formatting.
Tailored Feedback Dashboard
- Users receive a combined report showing:
  - Job Alignment Score: How well their resume matches their provided job description (e.g., “75% match”).
  - Top 3 Improvements:
    - Example: “The job description mentions ‘Python’ 12 times—add it to your ‘Skills’ section.”
  - Formatting Fixes: e.g., “Replace icons with text to avoid ATS parsing errors.”
Secure user Authentication:
- OAuth2 authentication allowing users to sign-in via Google or GitHub accounts with no password storage required.
- Session management: Automatic logout after a certain period of inactivity
Scalability:
- The system handles traffic surges during peak hiring periods (e.g., January, June) without delays.
- Users experience consistent performance even during busy hours.

Quality Attributes

Security

Security is the most critical quality attribute for JobFitAI. Resumes often contain personal details like phone numbers, addresses, and employment history, which must be protected to prevent misuse or identity theft. To achieve this, all uploaded resumes are encrypted during transfer and storage, ensuring that even if data is intercepted, it remains unreadable.

The system adheres to strict privacy laws, giving users full control over their data, and they can delete their resumes permanently. Access control ensures only authenticated users can view their own resumes. While encryption and compliance checks may slightly slow down processing, this trade-off is necessary to maintain user trust and safety.

Scalability

Traffic can surge by 300% during peak hiring seasons (e.g., January or graduation periods). Scalability ensures JobFitAI remains fast and reliable even when thousands of users access the platform simultaneously.

The system is designed to automatically adjust its resources based on demand, adding more servers when traffic spikes and reducing them during quieter periods. This flexibility prevents slowdowns or crashes when many users are active. A trade-off exists in higher operational costs during busy times, but this ensures the tool stays accessible when users need it most.

Testability

Testability ensures the AI’s feedback is accurate and helpful. The system is built to allow easy testing of new features and updates, ensuring changes don’t break existing functionality. For example, the AI’s suggestions are regularly compared with feedback from human career experts to verify their relevance. Automated checks run continuously to catch errors, like misreading resume sections or formatting. This level of testing and guarantees users receive reliable advice they can trust.

Evaluation Plan

Security Evaluation

To evaluate security, third-party penetration testers will simulate attacks such as SQL injection, cross-site scripting (XSS), and malicious file uploads. Vulnerabilities will be scored using Industry standard scales like CVSS scale, and critical issues will be resolved before launch. Compliance auditors will verify GDPR/CCPA adherence by checking data anonymization processes and encryption standards.

Scalability Evaluation

Scalability will be tested ito ensure the system handles real-world usage. Success criteria include:

9Simulate thousands of users uploading resumes and job descriptions at the same time. The system should stay responsive (e.g., no lag during morning rush hours).
Ensure the system adds resources automatically during busy periods and reduces them when idle.
Zero downtime during server failover tests (e.g., terminating 30% of EC2 instances).

Testability (AI Accuracy Evaluation)

Since manual HR feedback may not be feasible, we will:

Compare with Job Postings: Use real job descriptions (e.g., from LinkedIn) as “correct answers.” For example, if a job posting lists “Python,” we’ll check if the AI flags resumes missing this keyword.
User Validation: Ask users to rate suggestions (e.g., “Was feedback relevant to your target job?” on a 1–5 scale).
Historical Data Test: Run older “successful” resumes (from users who got jobs) through the AI to see if it identifies what made them effective.

User Feedback Evaluation

User satisfaction will be measured through:

Retention Rates: Track weekly active users (WAU) and resume re-uploads (indicates trust in the tool).
Surveys: Ask users, “Did JobFitAI’s suggestions improve your interview callback rate?”
“How likely are you to recommend JobFitAI to a friend?” (0–10 scale).