---
title: "Semantic search for jobs"
method: GET
path: "/search/job/embedding.json"
tags: ["Search"]
---

# Semantic search for jobs

`GET /search/job/embedding.json`

Harness the power of advanced AI embeddings to revolutionise how you search through job data. This endpoint transforms your search query into high-dimensional vector embeddings, then intelligently matches it against our entire job database using semantic similarity algorithms.

How it works:
1. AI Query Understanding - Your search terms are processed through neural embedding models that understand context, intent, and meaning
2. Vector-Based Matching - The system compares your query against vector representations of all job content in real-time
3. Intelligent Ranking - Returns results ranked by semantic similarity, not just keyword matching

Why this matters:
- Find jobs about "plumbing repairs" even when searching for "fixing pipes"
- Discover relevant work orders that use different terminology but share the same intent
- Uncover hidden patterns and connections in your job data that traditional search would miss

This isn't just search—it's AI that truly understands what you're looking for and delivers the most relevant results, even when the exact words don't match.

## Query parameters

- `q` string, required
- `limit` integer
- `similarity_threshold` number, float

## Response `200`

Embedding search results

- EmbeddingSearchResponse
  - `results` EmbeddingSearchResult[], required
    - `uuid` string, uuid, required — UUID of the found job
    - `type` string, required — Type of the object
    - `title` string, required — Title of the job
    - `description` string — Job description
    - `status` string — Current job status
    - `similarity_score` number, float, required — Similarity score between 0.0 and 1.0
    - `matched_content` string — The content that was matched in the embedding search
  - `query` string, required — The search query that was used
  - `count` integer, required — Number of results returned
  - `searchType` string, required — Type of search performed

## Other responses

- `400` — Bad request
- `500` — Internal server error
- `503` — Service unavailable - Embedding search not available

---

[API](https://skmtc.net/servicem8/apis/servicem8-api.md) · [All operations](https://skmtc.net/servicem8/apis/servicem8-api/llms.txt) · [OpenAPI document](https://skmtc-service-staging.skmtc.workers.dev/v1/apis/servicem8/servicem8-api/revisions/53ede60e3653/schema)