Skip to main content
Glama
Sign In

aitrainer.work AI training jobs

by work.aitrainer
OverviewDiscussionsAdmin

Search AI training jobs

search_jobs
Read-onlyIdempotent

Search open AI training / AI evaluation roles (RLHF, model rating, expert data work) across 20+ platforms by keyword and filters. Returns title, platform, pay, eligible countries and the aitrainer.work page for each role, most relevant first (newest first when there is no keyword). Every keyword must appear in the role, so use one to three words.

Input Schema

NameRequiredDescriptionDefault
limitNoResults to return, default 10.
queryNoWords to match in the title or description, e.g. "physics", "Spanish", "lawyer".
domainNoField of work.
countryNoOnly roles open to this country (English name, e.g. "India", "United States").
platformNoPlatform key, e.g. mercor, micro1, outlier, handshake, turing, mindrift. See list_platforms.
min_hourly_payNoMinimum hourly pay in USD (hourly roles only).

Output Schema

NameRequiredDescriptionDefault
jobsYes
returnedYes
browse_allYes
attributionYes
total_matchingYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • changedOutput schema / properties / jobs / items / properties / pay / description
      Previous value: -"Stated pay; \"(estimate…)\" marks an estimate, not the listing rate."New value: +"Pay as the listing states it. Text ending in \"(estimate, not stated in the listing)\" is our estimate; \"(platform range, not this role's rate)\" is the range the platform posts for all its projects."
  2. Changed1 schema field changed
    • changedOutput schema / (root)
      Previous value: -nullNew value: +{
+  "properties": {
+    "attribution": {
+      "type": "string"
+    },
+    "browse_all": {
+      "type": "string"
+    },
+    "jobs": {
+      "items": {
+        "properties": {
+          "match_reasons": {
+            "items": {
+              "type": "string"
+            },
+            "type": "array"
+          },
+          "pay": {
+            "description": "Stated pay; \"(estimate…)\" marks an estimate, not the listing rate.",
+            "type": [
+              "string",
+              "null"
+            ]
+          },
+          "platform": {
+            "type": "string"
+          },
+          "platform_name": {
+            "type": "string"
+          },
+          "slug": {
+            "type": "string"
+          },
+          "title": {
+            "type": "string"
+          },
+          "url": {
+            "description": "The role page on aitrainer.work, with the full listing and the application link.",
+            "type": "string"
+          }
+        },
+        "required": [
+          "title",
+          "platform",
+          "platform_name",
+          "slug",
+          "pay",
+          "url"
+        ],
+        "type": "object"
+      },
+      "type": "array"
+    },
+    "returned": {
+      "type": "integer"
+    },
+    "total_matching": {
+      "type": "integer"
+    }
+  },
+  "required": [
+    "total_matching",
+    "returned",
+    "browse_all",
+    "jobs",
+    "attribution"
+  ],
+  "type": "object"
+}
  3. First observed

TDQS

A4.5/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already cover the read-only, idempotent, non-destructive safety profile, and the description adds real behavioral context beyond them: AND-matching across keywords, most-relevant-first ordering with a documented tiebreak (newest first when no keyword), and the fields each result carries. It does not mention pagination beyond the limit parameter or any freshness/index staleness caveat.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Three sentences, front-loaded with capability, then return shape, then the one non-obvious calling rule. No filler and nothing repeated from the schema verbatim.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Six optional parameters, an output schema that defines the return shape, and full annotation coverage mean little is left to explain. The description supplies exactly the missing pieces: what corpus is searched, how ranking works, and how to phrase a query.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so baseline is 3; the description earns above that by explaining the non-obvious matching semantics of `query` ('Every keyword must appear'), which constrains how a caller should formulate terms. Other parameters (country, min_hourly_pay, domain) are left entirely to the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb (search) and resource (open AI training / AI evaluation roles) with the domain spelled out (RLHF, model rating, expert data work) and the corpus scope (20+ platforms). An agent can distinguish it from get_job (single record), match_jobs_to_profile (profile-driven), and list_platforms without opening any schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives concrete query guidance ('use one to three words') and routes the agent to list_platforms for platform keys. It stops short of explicitly stating when to prefer match_jobs_to_profile over keyword search, so the alternative-selection rule is implied rather than stated.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All aitrainer.work AI training jobs toolsInspect
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources