For developers and AI agents

Use Austin AI Live data

Everything Austin AI Live knows about AI in Austin is open to machines: a JSON API, an MCP server for AI assistants, and plain-text summaries. Free, read-only, no key.

The rules

Read-only. Nothing here can change data.
Only verified companies with an Austin office, or based in Austin or with a team here, are included.
Responses are cached for about 5 minutes; each carries a generated_at time.
Please cite Austin AI Live when you use the data: link the company's page (its url field, like https://austinai.live/c/apptronik) or https://austinai.live.

JSON API

Base URL https://austinai.live/api/v1. Full schema: /openapi.json.

GET /overview: Headline counts: companies by tier and layer, Austin HQs, new this week, events, jobs.
GET /companies: Companies on the map, A to Z. Filters: tier (native, powered), layer (builds, applies), hq (true, false), new (true), q (free text). on_map=false lists companies Austin AI Live tracks only for their AI jobs (on_map=all: both).
GET /companies/{slug}: One company with its Austin offices. 404 if it isn't listed.
GET /events: Upcoming Austin AI events from austinai.events, soonest first. Filter: company (slug).
GET /jobs: Jobs on the ATX AI Jobs board, newest first. Open only unless status=closed or all. view=all: AI roles at any Austin company (default: jobs at AI companies on the map). Filters (same as the board): company, role, seniority, workplace, industry, tier, layer, skill, pay, hq.
GET /jobs/{id}: One job, open or closed. 404 if it isn't on the board.

Lists page with limit (default 50, max 200) and offset, and include total and a next link. Bad parameters return 400 with a message. CORS is open, so browsers can call it directly.

curl "https://austinai.live/api/v1/companies?tier=native&hq=true"

MCP server

MCP (Model Context Protocol) is the standard way AI assistants call outside tools. Point any MCP client at https://austinai.live/mcp (Streamable HTTP, no login).

search_companies: List or search companies; same filters as /companies.
get_company: One company by slug.
list_upcoming_events: Upcoming events, optionally for one company.
list_jobs: Jobs on the board; same filters as /jobs.
get_job: One job by id.
austin_ai_overview: Headline counts for the whole ecosystem.

Limits: up to 50 results per call (page with offset), about 60 calls a minute per visitor, and answers may be up to 5 minutes old. Status and usage totals: /api/v1/status.

Claude Code:

claude mcp add --transport http austin-ai https://austinai.live/mcp

Claude or ChatGPT apps: add a custom connector with the URL above.

Discovery files

/robots: The front door for AI agents: every entry point, MCP setup, and the rules (also as /robots.md).
/llms.txt: What Austin AI Live is and every company, in Markdown for AI assistants.
/llms-full.txt: The full directory: descriptions, classifications, offices.
/openapi.json: OpenAPI 3.1 description of the JSON API.
/.well-known/api-catalog: Standard (RFC 9727) pointer to the API and its docs.
/sitemap.xml: Every page, including each company page.
/robots.txt: Crawlers welcome, AI crawlers included.

Company pages and the companies list also carry schema.org structured data (Organization, Place), and open job pages carry JobPosting.

What the fields mean

ai_tier: native: Would not exist without AI. powered: AI is real and material, but the business is not AI.
ai_layer: builds: Sells AI itself: models, chips, tools, infrastructure. applies: Sells an outcome that AI powers.
austin_hq: Headquartered in Austin. False means an Austin office or team with HQ elsewhere.
has_public_office: True when an Austin office is listed, with an address and a pin on the map. False for a company based in Austin, or with a team here, that publishes no office address: no pin, and offices is empty.
austin_since_year: The year the company came to Austin.
listed_at, new_this_week: When Austin AI Live first listed the company (not when it came to Austin).
slug: Stable id. It never changes once a company is listed.