owenkaplinsky
Clean initial commit for HuggingFace
363cda9
|
raw
history blame
2.27 kB

📄 CV Upload Stage — Overview

🎯 Purpose

The CV Upload Stage is the entry point of the HR screening pipeline.
It collects candidate information, stores the uploaded CV, and registers the applicant in the central database.
This stage serves as the bridge between the candidate and the agentic system,
initiating the automated recruitment workflow.


🧩 Role in the Pipeline

Step Description Responsible Component
1️⃣ Candidate applies Candidate fills out form and uploads CV via the web interface Streamlit UI
2️⃣ File saved CV file is stored in the shared volume (/src/database/cvs/uploads) File handler (save_cv)
3️⃣ Candidate registered Candidate details are inserted into the database DB client (register_candidate)
4️⃣ Status initialized Candidate record created with status = "applied" Database (candidates table)

🧱 Data Created

A new record is added to the candidates table containing:

Field Description
id Unique candidate UUID
full_name Candidate’s full name
email Unique email address
phone_number Optional contact number
cv_file_path Path to the uploaded CV file
parsed_cv_json Empty at this stage
status "applied"
created_at / updated_at Automatic timestamps

🧠 Next Stage Triggered

Once a record exists with status="applied",
the CV Parsing Agent is automatically triggered to extract structured data
(e.g., skills, education, and experience) from the uploaded CV.
This marks the transition toward the CV Screening phase.


✅ Key Takeaways

  • Initial entrypoint: introduces candidates into the system.
  • File persistence: all uploaded CVs are stored locally for later parsing.
  • Automation-ready: triggers the next agent without human input.
  • Simple interface: Streamlit UI makes the process candidate-friendly.

In short:
The CV Upload Stage is the gateway to the pipeline — it collects, stores, and initializes candidate data so the agentic workflow can proceed autonomously.