AI/myAi

Files

T

Build and Push Docker Images / build (push) Failing after 1m36s

Details

Add internet job search feature (cv-search-job)

- New cv-search-models shared library: EF entities + CvSearchDbContext for cvSearch schema (JobSearchTokens, JobSearchSessions, JobSearchResults tables)
- New cv-search-job worker service: polls DB for pending sessions, scrapes job boards via configurable HTML scraping, runs LLM scoring via cv-matcher-api, emails ranked results
- cv-matcher-api: JobTokenService creates one-time tokens; JobSearchController handles link clicks and creates sessions
- api: proxies job-search start endpoint, appends job search link to match result email
- CI workflow updated to build and push myai-cv-search-job:staging image
- CLAUDE.md documentation added for all affected services

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

2026-05-22 17:56:23 +03:00

2.7 KiB

Raw Blame History

cv-matcher-api — Internal CV Match Engine

Internal port 8082. Only reachable from api and cv-search-job via X-Internal-Api-Key.

Responsibilities

Indexes CV PDFs into the RAG system via rag-api
Matches a CV against a job posting URL (scrapes job HTML, scores pair with LLM)
Manages job search tokens and sessions for the one-click job search feature
Owns two EF DbContexts: CvMatcherDbContext (schema cvMatcher) and CvSearchDbContext (schema cvSearch)
Runs EF migrations for both contexts on startup

Key routes

Method	Route	Description
POST	`/api/cv/upload`	Index CV PDF into RAG
POST	`/api/cv/match-job`	Score CV against a job URL (LLM call)
POST	`/api/cv/find-jobs`	Find matching jobs from the RAG index
POST	`/api/cv/job-search/token`	Create a job search token (called by api after a match)
POST	`/api/cv/job-search/token/{tokenId}/start`	Validate token, create Pending session (called by api on link click)
GET	`/api/health`	Health check

Core services

CvMatcherService — orchestrates upload + match; calls IRagApiClient and IMatcherAiClient
JobTextExtractor — fetches a job page URL and extracts plain text
JobTokenService — creates tokens; validates + starts job search sessions; extracts CV keywords using simple heuristics (first 5 meaningful non-empty lines of CV text, split into words)

AI providers

Configured under Ai:Provider (OpenAI or Ollama). Both providers implement IMatcherAiClient.
Default model: gpt-4o-mini. Timeout: 90 s.

Database contexts

Both contexts use the same SQL Server connection string (from Database:* settings).

CvMatcherDbContext — schema cvMatcher; migrations in cv-matcher-api assembly
CvSearchDbContext — schema cvSearch; migrations in cv-search-models assembly (MigrationsAssembly = "cv-search-models")

Keyword extraction (JobTokenService.ExtractKeywords)

No LLM call. Takes the first 5 non-empty lines of CV text that are:

Longer than 5 characters
Not purely numeric or contact-line patterns

Splits into words, strips punctuation, deduplicates, returns up to 10 comma-separated keywords.
These keywords are stored in JobSearchSessionEntity.Keywords and used by cv-search-job for scraping.

Settings

Section	Notes
`Database`	Shared SQL Server connection
`RagApi`	BaseUrl + InternalApiKey for rag-api
`Ai`	Provider, model, timeout
`Matcher`	TopK, DeepScoreTopN, MaxJobTextChars
`JobSearch`	TokenExpiryDays, providers list (stored in session JSON)
`InternalApi`	ApiKey used by UseInternalApiKeyProtection middleware

2.7 KiB Raw Blame History