Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/mauroperez055/infoJobs/llms.txt

Use this file to discover all available pages before exploring further.

The GET /ai/summary/:id endpoint generates a concise, human-friendly summary of a job listing using a locally running Ollama language model. The response is streamed back to the client as plain text chunks so that the UI can progressively display content as it is produced — rather than waiting for the full generation to complete. The summary is written in Spanish, is 4–6 sentences long, and covers the role, company, location, and key requirements. 
GET /ai/summary/:id
Ollama must be running locally at http://localhost:11434 for this endpoint to function. The model used is qwen2.5:3b. If Ollama is not running or the model is not available, the endpoint returns a 500 error (provided headers have not yet been sent).

​
Path Parameters

​
id
string
required
The UUID of the job listing to summarize. The job must already exist in the store.

​
Response Format

This endpoint does not return JSON. The response is a raw text stream delivered with the following headers:
HeaderValue
Content-Typetext/plain; charset=utf-8
Transfer-Encodingchunked
The response body is a stream of Markdown-formatted text chunks. Clients should consume the stream incrementally — each chunk is a fragment of the final summary as produced by the model.

​
Rate Limiting

To prevent abuse of the local AI resource, this endpoint is rate-limited using express-rate-limit with the draft-8 standard headers.
SettingValue
Window60 seconds
Max requests per IP5
HeadersRateLimit (draft-8 spec)
When the limit is exceeded the server responds immediately with HTTP 429: 
{ "error": "Demasiadas solicitudes, por favor intenta de nuevo más tarde." }

​
Example

curl http://localhost:1234/ai/summary/7a4d1d8b-1e45-4d8c-9f1a-8c2f9a9121a4
The response streams plain-text Markdown directly to stdout: 
**Desarrollador de Software Senior en Tech Solutions Inc.**

Tech Solutions Inc. busca un Desarrollador de Software Senior con sede en Buenos Aires, Argentina
en modalidad híbrida. El rol requiere experiencia sólida en Node.js, TypeScript y PostgreSQL,
con al menos 5 años en el desarrollo de APIs REST escalables. El candidato ideal tiene
experiencia liderando equipos y tomando decisiones de arquitectura...

​
Frontend Integration

The React frontend consumes this endpoint via a custom useAISummary hook that manages the fetch lifecycle, accumulates streamed chunks, and exposes loading and error states: 
const { summary, loading, error, generateSummary } = useAISummary(jobId);
ValueTypeDescription
summarystringAccumulated Markdown text received so far from the stream
loadingbooleantrue while the stream is in progress
errorstring | nullError message if the request failed, otherwise null
generateSummaryfunctionCall this to initiate the summary request

​
Error Responses

​
404 — Job Not Found
object
Returned when the provided job ID does not match any record in the store. Headers have not yet been sent, so the response is standard JSON.
{ "error": "Job Not Found" }
​
500 — Generation Error
object
Returned when Ollama is unavailable or throws an error before any content has been streamed. If content was already streaming when the error occurs, the connection is closed silently via res.end().
{ "error": "Error generating summary" }

Build docs developers (and LLMs) love

Get started for freeTalk to us