Cognition (Devin and Windsurf)

1. 2024-09

   Devin ARR $1M

   [3]

2. 2025-01-08

   Answer.AI independent evaluation: 3 successes, 14 failures, 3 unclear out of 20 tasks

   [6]

3. 2025-06

   Devin ARR $73M

   [3]

4. 2025-07-14

   Windsurf acquisition

   "$82M ARR", "350+ enterprise customers", enterprise ARR "doubling quarter-over-quarter"

   [1]

5. 2026-05-27

   $1B+ round at $26B (blog title); $492M annualized run-rate (secondary data)

   [7]

6. 2026-09-25

   "Cognition Crosses $1B in Annualized Revenue Run Rate" (blog title; text could not be opened)

   [5]

Acquisition channels

Three channels. First, inorganic: the Windsurf acquisition doubled ARR in one move; the company blog says "ARR more than doubled" with the deal [3]. Second, enterprise sales and a branded customer list: Goldman Sachs, Citi, Dell, Cisco, Ramp, Palantir, Nubank, Mercado Libre [3]. Third, a lower self-serve price: the Core plan starting at $20 in April 2025 [4].

Loop

Enterprise pilot, Devin merging PRs, usage data, a lower cost per task, more tasks. According to the founder, Devin merges 30 to 40 percent of PRs at some customers (secondary data) [8]. The jump from $1M to $73M ARR (September 2024 to June 2025) overlaps with the Core plan period [3]; the causality is not a company claim (opinion).

Activation

Devin 2.0 moved the first value moment from a fully autonomous task to planning and search: "Interactive Planning", Devin Search, Devin Wiki [4]. The user sees value before the agent finishes a task (opinion).

Retention evidence

No retention evidence. The Windsurf side's "we are doubling our enterprise revenue every quarter" [2] is an expansion signal, not retention. Churn, NRR or task success rate were never published.

Role of pricing

Self-serve gate at $20 [4]; enterprise revenue by contract. In September 2025 the company wrote that total net burn across its whole history was under $20M [3]; burn after the 2026 rounds was not disclosed.

Distribution moves

Windsurf is described as a "scaled GTM machine" [3]; the sales team was one reason for the deal (opinion). The speed of the acquisition in the founder's telling: news on Friday evening, signature on Monday morning; "if there is something to do here at all, then it has to be ready to go by Monday morning" (secondary data) [8]. Customer overlap under 5 percent [3].

Team

About 35 people before Windsurf, 21 of them former founders (secondary data) [8]. All Windsurf employees joined the deal and vesting cliffs were waived [1].

The five questions of the Growth Claim Check, applied to this company's public claims.

Test your own claim with the same five questions: Growth Claim Check →

• Answer.AI's month-long independent test (January 2025): "Tasks it can do are those so small and well-defined that I may as well do them myself, faster, my way." They ended up recommending developer-driven tools [6].
• Windsurf moved to Cognition a week after losing its founders and research team; Jeff Wang's words: "Last week, we lost our founders and our research team." [2] The detail of the move to Google was not opened in a primary source (secondary data).
• The valuation went from $10.2B to $48B in a year [3][5]; the revenue definition did not get clearer at the same pace (opinion).

This section is opinion; a reading, not a source.

1. 01

   (opinion) Acquisition is a growth channel; but next to "ARR doubled" write the line "how much was organic" yourself.

2. 02

   (opinion) Lowering the price gate opens the product to trial; enterprise revenue still comes through a sales team. Do not treat the two as one funnel.

3. 03

   (opinion) A product can be weak in independent tests and strong in revenue; the buyer and the user are different people. Measure "who approves, who uses" in your own product.

4. 04

   (opinion) Shipping your own small model is a margin decision. With dollar costs and lira revenue, that question reaches you earlier.

5. 05

   (opinion) Small team, high authority: a team made mostly of former founders is a hiring thesis, not decoration.

A question for you: remove this company's logo and drop the loop onto your own product. Which node stays empty?

Numbers are the references in the text. An 'opened' source was read by the researcher; 'secondary' ones are verified against their primary source before publication.

1. [1]
   Cognition's acquisition of Windsurf

   Cognition (blog) · 2025-07-14 · opened

   “$82M ARR · 350+ enterprise customers · doubling quarter-over-quarter”

2. [2]
   The Next Chapter (statement from the Windsurf side)

   Windsurf / Jeff Wang · 2025-07-14 · opened

   “Last week, we lost our founders and our research team.”

3. [3]
   Funding, growth, and the next frontier of AI coding agents

   Cognition (blog) · 2025-09-08 · opened

   “ARR $1M (September 2024) → $73M (June 2025) · Total net burn under $20M across entire history”

4. [4]
   Devin 2.0

   Cognition (blog) · 2025-04-03 · opened

   “a new Core plan starting at $20”

5. [5]
   Cognition Crosses $1B in Annualized Revenue Run Rate

   Cognition · accessed 2026-10-05 · opened

   “Today, Cognition crossed $1B in annualized revenue run rate.”

6. [6]
   Thoughts on a Month with Devin

   Answer.AI (Hamel Husain, Isaac Flath, Johno Whitaker) · 2025-01-08 · opened

   “Tasks it can do are those so small and well-defined that I may as well do them myself, faster, my way.”

7. [7]
   AI coding startup Cognition raises $1B at $25B pre-money valuation

   TechCrunch · 2026-05-27 · secondary

   Relayed wording: $492 million annualized run-rate

   Note: Secondary data: news relay; the page was not opened.

8. [8]
   Cognition CEO Scott Wu on acquiring Windsurf (written summary)

   Cheeky Pint (Stripe podcast) · 2025-08-27 · secondary

   Relayed wording: if there is something to do here at all, then it has to be ready to go by Monday morning

9. [9]
   Scott Wu on Acquiring Windsurf (episode page)

   20VC · 2025-07-18 · secondary

   Note: episode description only; no transcript

• Text of the May and September 2026 blog posts: run-rate definition, customer count, cash burn.
• Devin's older $500/month plan; churn and talent loss after Windsurf.
• Revenue split between Devin and Windsurf; self-serve versus enterprise.
• Full transcripts of the Cheeky Pint and 20VC conversations.