mdasar2927/employ-ai

Purpose and entry points

Employ-AI is a career-support application organized as static, task-specific frontend screens plus a Python backend API. User workflows cover profile creation, resume handling, skill analysis/gaps, internship recommendations, learning, and reporting.

Primary entry points:

• Frontend landing screen: 1. main screen/main_screen.html

• Backend application bootstrap: backend/app/main.py

• Shared browser API/auth integration: shared/api.js, shared/auth.js

Frontend workflow surfaces

The frontend is split into independent HTML screens rather than a single bundled web application. Each screen represents a major user-facing stage and likely calls the common shared API/auth layer.

Core screens:

• Profile: 2. profile/profile.html

• Resume: 3. resume/resume.html

• Skill analysis and gap views: 4. skill analysis/skill_analysis.html, 6. skill gap/skill_gap.html

• Internship recommendation: 5.internship recommendation/internship_recc.html

• Learning and report outputs: 7. learner/learner.html, 8. report/report.html

INTEGRATION_MAP.md is the likely contract map between these screens and backend endpoints.

API boundary and request organization

The backend is structured around routers, schemas, repositories, and services. Routers expose feature APIs; schemas define request/response models; repositories isolate persistence access; services hold analysis and generation logic.

Feature API modules:

• backend/app/routers/profile.py

• backend/app/routers/resume.py

• backend/app/routers/matching.py

• backend/app/routers/internships.py

• backend/app/routers/learning.py

• backend/app/routers/reports.py

Corresponding typed contracts live in backend/app/schemas/, including backend/app/schemas/profile.py, backend/app/schemas/resume.py, and backend/app/schemas/matching.py.

Career analysis pipeline

The principal control flow is profile/resume input into analysis, matching, learning-roadmap, and report outputs.

Key processing components:

• Resume parsing: backend/app/services/resume_parser.py

• AI-backed analysis: backend/app/services/ai_analyzer.py

• Candidate-to-opportunity matching: backend/app/services/matching_engine.py

• Learning/roadmap generation: backend/app/services/roadmap_generator.py

Likely flow: the resume and profile APIs persist user inputs, parsing/AI analysis derives skills or candidate attributes, matching evaluates internship fit, roadmap generation converts gaps into learning recommendations, and reporting aggregates these results.

Persistence and security boundary

Supabase is the explicit database integration and is accessed through a dedicated database module:

• backend/app/db/supabase.py

• backend/schema.sql

• backend/migrations/001_rls_security.sql

Repositories form the persistence boundary for domain data:

• backend/app/repositories/profiles.py

• backend/app/repositories/resumes.py

• backend/app/repositories/matches.py

• backend/app/repositories/internships.py

• backend/app/repositories/learning.py

• backend/app/repositories/reports.py

Row-level security is architecturally material because backend/migrations/001_rls_security.sql indicates database-enforced tenant/user access controls. Frontend authentication handling is centralized in shared/auth.js.

Runtime and configuration

The backend is a Python service with dependency and configuration layers:

• backend/requirements.txt

• backend/app/core/config.py

• backend/app/api/dependencies.py

• .env.example

Environment configuration should supply Supabase connectivity and any external AI-provider credentials consumed by backend/app/services/ai_analyzer.py. The frontend remains static HTML/JavaScript, using shared/api.js as its backend-service boundary.