# Carrot Labs: Usage-based attribution for every AI dollar

Canonical: https://mudpie.ai/companies/carrot-labs/
Breadcrumb: [Home](https://mudpie.ai/) / [Companies](https://mudpie.ai/companies/) / [Carrot Labs: Usage-based attribution for every AI dollar](https://mudpie.ai/companies/carrot-labs/)
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.

Carrot Labs, whose current product is branded SuperPenguin, helps teams see and attribute AI spend across providers. It fits a company whose model bill is growing faster than its ability to explain which customer, feature, team or pull request caused it.

## What it does

[SuperPenguin’s current homepage](https://superpenguin.ai/) connects provider usage across OpenAI, Anthropic, Gemini, Deepgram, ElevenLabs, Bedrock, Azure, Modal, Cursor, OpenRouter and others. Its SDKs tag requests by customer, feature, team or environment; its coding ROI surface attributes AI spend to pull requests; and alerts can go to Slack, email or Discord. The docs say traffic goes directly to the provider—the SDK observes calls rather than acting as a proxy.

The useful decision is whether the company needs attribution or merely a monthly total. A finance or product team can ask which customers are unprofitable after inference cost, which feature is burning an Anthropic budget or how much an AI-written PR cost to ship. The attribution dimension is only as good as the metadata the engineering team adds, so the first pilot should attach costs to one product or repository.

Pricing is clear and based on managed AI spend rather than seats or requests. [The pricing page](https://superpenguin.ai/pricing) lists Free at $0 up to $2K managed spend, Growth at $30/month up to $5K, Pro at $200/month up to $20K and Enterprise custom. Free includes SDK attribution, coding ROI for the user’s own spend and a provider-wide view; Growth adds team members, alerts and backfill; Pro adds cost per merged PR and forecasts.

## Founder context and tradeoffs

The YC profile identifies Christopher Acker and Yuta Baba as founders. Yuta’s public background includes data science at Snowflake and account-based bookings forecasts; the launch describes Christopher’s AI work at Skylo. That maps to spend attribution and financial planning.

The buyer should ask how provider invoices reconcile with SDK estimates, how missing metadata is handled, what historical backfill covers and whether coding-cost data includes all relevant agents. A dashboard can explain the bill without lowering it; the savings decision still belongs to the team.

## Editorial take

I would shortlist SuperPenguin when AI cost is a product-margin or engineering-budget problem, not just a curiosity. Start with one provider and one attribution dimension. If the company cannot enforce metadata at the call boundary, the extra dashboard will mostly rearrange the same unknown total.

## Quick facts

| Field | Sourced detail |
| --- | --- |
| Product | AI cost attribution, coding ROI and provider-wide spend management |
| Buyers | AI product, finance, FinOps and engineering teams |
| Public pricing | Free $0/$2K spend; Growth $30/$5K; Pro $200/$20K; Enterprise custom |
| Integrations | Multiple model, voice, coding and inference providers listed |
| Main gate | SDK coverage, metadata quality, invoice reconciliation and privacy |

## Sources checked

| Source | Checked |
| --- | --- |
| [YC profile](https://www.ycombinator.com/companies/carrot-labs) | 2026-09-19 |
| [SuperPenguin homepage](https://superpenguin.ai/) | 2026-09-19 |
| [SuperPenguin pricing](https://superpenguin.ai/pricing) | 2026-09-19 |
| [SuperPenguin docs](https://superpenguin.ai/docs) | 2026-09-19 |

## Cohort context

Carrot Labs 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 2026-09-19T16:20:06.156Z 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 | Observed |
| 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.