Breaking
YC F25 — Parrot launches on the App Store: TikTok for Spanish Method — Comprehensible input, first proposed by Dr. Stephen Krashen Signal — 60% of Spanish 101 students would pay $30 for early access Team — CTO Erik Dahl self-taught 3 languages in one year Homework — 52 apps analyzed, 65 learners interviewed pre-launch Usage — Power users spend ~6 hours a week in the feed YC F25 — Parrot launches on the App Store: TikTok for Spanish Method — Comprehensible input, first proposed by Dr. Stephen Krashen Signal — 60% of Spanish 101 students would pay $30 for early access Team — CTO Erik Dahl self-taught 3 languages in one year Homework — 52 apps analyzed, 65 learners interviewed pre-launch Usage — Power users spend ~6 hours a week in the feed
Company · Edtech · Consumer · San Francisco

Parrot Wants You to Learn Spanish the Same Way You Waste Time

The Y Combinator (F25) app looks like TikTok and behaves like immersion. Swipe through native videos, tap any word, and let the feed do the teaching. The bet: understanding beats drilling.

Open the app and the first thing you notice is what is missing. No cartoon owl guilt-tripping you about your streak. No matching pairs of words to a robotic voice. Just a video: a person, speaking Spanish, at normal speed, about something ordinary. You do not understand every word. You understand enough. You swipe. That, more or less, is the entire pitch of Parrot.

Parrot is a language-learning app out of Y Combinator's Fall 2025 batch, built by three founders in San Francisco. Its shorthand is "TikTok for language learning," which is accurate enough to be useful and simple enough to be slightly misleading. The format is a vertical, swipeable feed of short native-speaker videos. The method underneath it is a decades-old idea from second-language research called comprehensible input.

For now, it teaches one language: Spanish. The app is a free download on the App Store, with a subscription for full access.

The problem with staying on a streak


The gap Parrot noticedBefore writing code, the founders did something unfashionable: homework. They analyzed 52 language-learning apps and interviewed 65 learners on Reddit and Discord. The conclusion they came to is one many people quietly suspect - the popular apps are very good at engagement and revenue, and far less good at producing people who can actually hold a conversation.

This is not a small distinction. Engagement and fluency are two different products. One keeps you tapping every day. The other lets you order dinner in Mexico City without freezing. Parrot's argument is that the industry optimized for the first and quietly gave up on the second.

The founders' verdict after studying the field: the apps make money and streaks. They do not reliably make speakers. Paraphrased from Parrot's YC launch

How the feed actually teaches


The theory Parrot leans on comes from Dr. Stephen Krashen, whose comprehensible-input hypothesis dates to the 1980s. The claim is straightforward: you acquire a language by understanding messages slightly above your current level, not by memorizing rules and reciting them. It is closer to how a child picks up a first language than to how most of us suffered through high-school conjugation tables.

Parrot's job is to package that theory into a phone. Here is the loop:

1

Watch

A native-speaker video at real speed, chosen to sit just above your level.

2

Tap

Interactive subtitles - tap any word for an instant translation.

3

Save

New words drop into a personal dictionary for spaced-repetition review.

4

Speak

Parrot-back practice and an AI tutor give feedback instead of drills.

Around that core sit the scaffolding pieces: a "Flight School" beginner track that sequences early lessons, a grammar library covering more than 250 topics for when you want the rule explained, and a personalized algorithm that decides what to show you next. The point of the extras is that you never have to leave the feed to look something up.

Where a learner's time goes (illustrative)
Real input
82%
Drills
10%
Grammar ref
8%
A rough sketch of Parrot's design bias: watching real speech is the lesson; grammar and review are the support cast, not the main event.

Who it is for


The target user is anyone who has bounced off traditional apps: complete beginners through intermediate learners, busy professionals who can spare ten to thirty minutes a day, travelers who want practical Spanish, and students who already live inside a video feed. Parrot's not-so-subtle framing is that it wants to replace the dorm-room scroll - to point the same restless attention at something that compounds.

The early demand signal is worth pausing on. When the founders showed an MVP prototype to a Spanish 101 class, 60% of the students said they would pay $30 for early access. Students are the most broke and most skeptical customers imaginable. Getting them to reach for a wallet before the product exists is a real data point.

52Apps analyzed pre-launch
65Learners interviewed
~6 hrsPower-user time / week

The people building it


There are three co-founders. Amir Hanna is the CEO. Erik Dahl is the CTO, and his origin story is the tidy version of the whole thesis: he taught himself Spanish, French, and Italian in a single year, largely through immersion rather than any app. Julia Hudea is the CPO, with dual degrees in mechanical engineering and computer science. Two of the founders had previously worked together on a pharmacy voice assistant before turning to language.

A CTO who learned three languages in a year is now building the tool he wishes had existed while he did it.

Where it sits in a crowded market


Language learning is not an empty field. Duolingo owns the habit-and-gamification corner. Babbel, Rosetta Stone, Busuu, and Pimsleur split the more structured, course-like segment. Closer to Parrot's approach are video-immersion tools such as Lingopie and FluentU, which also lean on authentic content.

Parrot's wager is on format. Immersion tools have existed for years, but they often feel like a media library you have to work through. Parrot borrows the one interface that has proven it can hold human attention indefinitely - the infinite vertical feed - and puts comprehensible input inside it. The method is old. The delivery mechanism is the part that is new, and it is the part that could matter.

Positioning: how apps compete
Gamified
Duolingo
Course
Babbel
Video feed
Parrot
Illustrative map of approach, not market share. Parrot's differentiation is the feed-native format around an old immersion idea.

The business, and the guarantee


The model is a straightforward consumer subscription: free download, free trial, premium tier for full access. What is a little unusual is the confidence baked into the marketing. Parrot's site claims users can learn Spanish meaningfully faster with video-based comprehensible input, and dangles a money-back framing around reaching real conversations. Guarantees are a way of putting your method where your mouth is. They only make sense if you believe the method works.

It is early. The company raised a small seed alongside its YC batch, the app is live, and the roadmap points toward more languages beyond Spanish. Whether the feed can carry a learner all the way to fluency, or only to the comfortable middle where most apps quietly plateau, is the open question. Parrot has at least picked the right problem to be judged on.

Parrot at a glance
Founded
2024, San Francisco
Batch
Y Combinator, Fall 2025 (F25)
Founders
Amir Hanna (CEO), Erik Dahl (CTO), Julia Hudea (CPO)
Product
Parrot - Learn Spanish (iOS)
Method
Comprehensible input via short-form native video
Model
Freemium consumer subscription
#language-learning#spanish#comprehensible-input #tiktok-for-learning#edtech#yc-f25 #consumer-app#ai-tutor#duolingo-alternative