ATS PoC — Technical Exercise

Design and build a micro–product that demonstrates your ability to:

• Model a scalable, maintainable architecture (DDD + Hexagonal + CQRS / events).

• Apply good engineering practices (testing and code quality).

• Integrate asynchronous processing with messaging.

• Use AI to enrich real candidate data.

• Deliver a simple web UI to submit and explore applications.

---

Product Scope

Build a minimal Application Tracking System (ATS) with the following capabilities:

1) Apply to a Job

• A page with a job description where a candidate submits: full name, contact info (email/phone), position, and optional notes.

• A textarea to paste the CV as plain text (no file upload/OCR).

• On submission, store the application with appliedAt (auto) and initial status (e.g., received).

• Trigger an asynchronous AI enrichment that:

  • Produces a concise CV summary.

  • Generates a relevance score based on the position applied for.

2) Browse Applications

• A page listing applications newest first.

• Real-time filtering by status and position.

• Simple search (e.g., by candidate name or email).

• A detail view showing candidate data, pasted CV text, AI summary, AI score, status, and timestamps.

---

Functional Requirements

• Submit an application containing candidate data, position, and CV pasted as text.

• Persist the application and initiate the AI enrichment process asynchronously.

• Update the application with AI summary and AI score when enrichment is completed.

• List applications descending by appliedAt, with filters and search.

• Provide an application detail view.

---

Non-Functional Requirements

• Clear structure reflecting DDD + Hexagonal + CQRS/events.

• Unit and integration tests

• Clean, readable code and concise documentation.

• Straightforward local run instructions.

(Do not include implementation details in this document.)

---

UI Requirements

• Apply Page with inputs for candidate data and CV textarea.

• Applications Page with search + filters and newest-first ordering.

• Detail Page displaying original CV text, AI summary, AI score, status, and timestamps.

---

AI Enrichment (asynchronous)

• After an application is submitted, enqueue an enrichment task.

• The task summarizes the CV and scores relevance with respect to the selected position.

• The result becomes visible in the application detail and in the list (score column).

---

Acceptance Criteria

• Submitting an application creates a record with appliedAt and default status.

• Asynchronous enrichment adds summary and score to that application.

• Applications list is newest first and supports real-time filtering and search.

• Detail view shows all relevant data including enrichment outputs.

• Tests cover submission, filtering, and enrichment paths.

---

Deliverables

• Git repository with a README.md and source code.

• A make init command is desirable to start running the application.

---

Constraints

• CV input must be via textarea (plain text).

• No file uploads or OCR.

• Mock LLM requests, don't use a real API.

• Use any stack you prefer, as long as the requirements above are met.

---

Timebox & Submission

• Share a GitHub repository link with instructions to run and test locally.

---

Evaluation

• Clarity of domain boundaries and event flow.

• Correctness of the user flows and asynchronous enrichment.

• Code quality and test completeness.

• Simplicity and usability of the UI.

• Documentation sufficiency (without implementation detail).