# Autumn AI turns live public-web research into sourced people and company records

Canonical: https://mudpie.ai/companies/autumn-ai/
Breadcrumb: [Home](https://mudpie.ai/) / [Companies](https://mudpie.ai/companies/) / [Autumn AI turns live public-web research into sourced people and company records](https://mudpie.ai/companies/autumn-ai/)
Author: Ali Abouelatta (https://mudpie.ai/authors/ali-abouelatta/)
Published: 2026-09-19
Updated: 2026-09-19
Research type: Company profile
Method: Company and accelerator sources checked 2026-09-19. Product claims are attributed to their sources; this is research, not a hands-on product trial.

Autumn AI is for teams that need fresh, source-backed research on people and companies rather than another static database. Its product decision is unusually clear: pay for live research tasks and provenance, or keep buying rows from a conventional enrichment tool.

## What it does

Autumn says it resolves fragmented public-web information into profiles of people and companies, including relationships, work history, contact information and digital footprint. Its public examples cover cohort building, company research, incorporation monitoring, account trees and intent signals. [Autumn homepage](https://www.autumn.ai/)

The API and docs make the operating model concrete. A user submits a task with a goal, Autumn plans and searches public sources, cross-checks facts, then returns rows with source links. Tasks have a durable ID, streamed events, a sandbox and fetchable output, so a long research job can continue after the original request. [Autumn docs](https://www.autumn.ai/docs) That is useful for a founder who wants research embedded in a product or recurring GTM workflow, not only a one-off chat answer.

## Price and tradeoffs

The public plans are Hobby at $20 per month for 2,000 credits, Growth at $75 for 8,000 credits and Scale at $280 for 32,000 credits; Enterprise is custom. [Autumn pricing](https://www.autumn.ai/pricing) The credit model is straightforward enough to budget, but the real cost depends on how many sources and fields a task consumes. The team should define which claims need fresh research, which can be cached and how source disagreement is surfaced before turning it loose on a large prospect universe.

Autumn’s advantage over a static database is also its main risk. Matching records across filings, profiles, code, posts and news can uncover a new company earlier, but identity resolution and public-web freshness still need review. Contact data should be handled within the team’s legal and privacy rules; “source-backed” does not mean every field is correct or every use is appropriate.

The [YC profile](https://www.ycombinator.com/companies/autumn-ai) identifies Vishnu Sampathkumar and Shiv Kampani as co-founders and describes their ML, VC-sourcing and research backgrounds. That is relevant context for a data product, not a guarantee of match quality.

## My editorial take

I would choose Autumn when a team’s edge comes from discovering changing public signals—new incorporations, hiring, launches or account movement—and when each row needs an audit trail. I would not replace a stable CRM database with it blindly. Start with one repeatable research task, cap credit use and inspect a sample of matched identities before expanding the universe.

## Quick facts

| Field | Sourced detail |
| --- | --- |
| Buyer | Sales, recruiting, investing and risk teams |
| Product | Live public-web research, profiles and API tasks |
| Pricing | $20/2,000 credits, $75/8,000, $280/32,000; Enterprise custom |
| Data model | Task-based API with sourced output rows |
| Main fit question | Do changing public signals matter enough to justify live research? |

## Sources checked

Autumn’s YC profile, homepage, pricing page and API documentation were checked on 2026-09-19.

## Cohort context

Autumn AI is listed in Winter 2026. In our 2026-09-18 directory snapshot, 126 of 199 listed companies in that cohort have YC’s primary industry label B2B (63.3%). This is a current-directory comparison, not an original intake count or a performance ranking. [Nine-cohort dataset](https://mudpie.ai/research/yc-cohorts-2026-09-19.json).

## Public website snapshot

Observed 2026-09-19T16:20:02.187Z in raw homepage HTML. This records visible metadata and advertised links, not agent execution or product quality.

| Signal | Homepage observation |
| --- | --- |
| Product description metadata | Observed |
| Canonical link | Observed |
| H1 or H2 heading | Observed |
| Typed structured data | Observed |
| Docs/developer link | Observed |
| Pricing link | Observed |
| llms.txt link | Not observed in this response |
| Markdown alternate | Not observed in this response |

[Public observations](https://mudpie.ai/research/yc-homepage-links-2026-09-19.json) · [Collection method](https://mudpie.ai/research/yc-homepage-methods/README.md). Missing links here do not establish that a capability or file is absent elsewhere.


## Author disclosure

I cofound Lazyweb and publish Mudpie. This is an owner-written publication, not an independent testing organization. Research notes distinguish observations, sourced reporting and editorial judgment.