Learning and careers decision
Parlai
A small technical team (or single capable developer with coding assistants) can build a usable WhatsApp-based AI language tutor replacement; core pieces (WhatsApp webhook, LLM prompts, STT/TTS, DB) are reproducible and prior open-source projects cover voice/chat pieces.
Visit website↗$12/mo
$144/yr
Read off the official pricing page.
$100one-off44 h to build
$200/mo6 h/mo upkeep
On cash alone, building overtakes the subscription at 18 seats.
No open-source build does this yet
Nothing published replaces this one, so a replacement starts from an empty file. Here is what it would have to cover.
What a replacement has to do
- User sends text or voice message via WhatsApp → webhook receives message → run STT if voice → call LLM with prompt to continue conversation + produce corrections/feedback → store update to user session and send text/voice response back via WhatsApp API.
What it still won’t have
- Production-grade WhatsApp approvals, phone-number provisioning and scale handled by vendor
- Designed UX refinements, analytics and polished curriculum sequencing
- Enterprise features (company/academy onboarding, invoices, vouchers) and support ops
- GDPR-ready legal/DSAR flows and certified data-handling processes out of the box
- Referral reward system and payment/subscription portal integration
What remains hard
- Product polish and ongoing maintenance
First-year cost
Keep paying
Paying is—cheaper in year one.
On cash alone, building overtakes the subscription at 18 seats.
Money you would actually spend
Time you would spend
—
What you would spend
What we assumed
The verdict above measures whether you could build it. This one is only about money.
Runnable build prompt
Build a minimal Parlai replacement: Node.js (Express) backend, Postgres for user/session storage, Docker for deployment, and Redis for background jobs. Core features in scope: WhatsApp Cloud API webhook to receive/send text and voice notes; integrate OpenAI (or equivalent) for conversation + correction prompt pipeline; integrate Whisper (or speech-to-text API) for incoming voice notes and a cloud TTS for voice replies; persist user profiles, CEFR level, conversation history, and monthly minutes; implement simple routing to handle free vs premium quota; add unit and integration tests for webhooks, LLM calls, and STT/TTS flows; include error handling, retries, logging, and basic monitoring. Out of scope: enterprise admin portal, advanced curriculum authoring, payment provider UI, and GDPR-compliance certification. Deliverables: Docker-compose manifests, Terraform/DigitalOcean (or AWS) deploy instructions, OpenAPI for webhook endpoints, test suite, and runbook for WhatsApp number provisioning and secrets management.
How we checked
How the score was reached
- Partly verdict base52
- 3 cited sources+3
- Price verified on pricing page+3
- Evidence score58
The base comes from the verdict. Everything under it is a check that either happened or did not, and each one is a fact frozen in this record rather than a judgement made at render time - so the same evidence always produces the same number.
How scoring works →Cited sources · 3
Every page the run actually retrieved.
- official productParlai - Your personal AI language tutor on WhatsApp
- official pricingParlai pricing and plans
- official docsParlai product facts: AI language practice on WhatsApp
Integrity checks
What held up, and what did not.
