AI assistants and search decision

Alchie

A competent developer can reproduce the core interview→transcribe→LLM post-generation and scheduling flows in a few weeks, but the full product experience (strategy workflow, support channels, dataset-driven polish and reliability) would be costly to match.

Visit website
You pay

$19/mo

$228/yr

Read off the official pricing page.

You’d pay instead

$100one-off56 h to build

$60/mo6 h/mo upkeep

On cash alone, building overtakes the subscription at 4 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

  • Record a short interview, transcribe audio, analyze profile and sample posts to infer voice, generate a draft post and photo suggestions via LLM, allow user edits, and optionally schedule/publish to LinkedIn.

What it still won’t have

  • Alchie's claimed dataset and social proof ("1 500 posts rédigés", "20M d'impressions")
  • Built-in editorial strategy and calendar workflow included in plan Argent
  • Access to vendor-run support channels (group WhatsApp) and onboarding
  • Polished UX, moderation, and reliability of a production SaaS
  • Seamless Stripe-managed trial/renewal flows and billing polish

What remains hard

  • Product polish and ongoing maintenance
Read the build prompt

First-year cost

Keep paying

Paying is—cheaper in year one.

On cash alone, building overtakes the subscription at 4 seats.

Paid seatsseats

Money you would actually spend

Keep paying
—

Subscription price × seats × 12

Build it
—

AI build —APIs + hosting —

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

Not run yet
Build a minimal Alchie-like web app using Next.js (React) + Postgres + Vercel (or small VPS) that: 1) records or uploads short audio interviews and stores files in S3; 2) runs STT (use Whisper or a managed STT API) and saves transcriptions; 3) analyzes user profile and up to 3 example posts to extract voice/style (embeddings + simple heuristic); 4) composes prompts and calls an LLM API to generate a LinkedIn post draft, two photo-idea strings, and three short storytelling tips; 5) provides an editor UI to edit drafts, request retouches (counted), and save to a user library; 6) supports LinkedIn OAuth and schedule/publish via LinkedIn API; 7) implements Stripe subscription checkout for single-seat Bronze plan; explicitly out of scope: advanced editorial calendar engine, WhatsApp group support, large-scale analytics, and marketing site polish. Include error handling for failed STT/LLM calls, retries, input validation, basic automated tests for STT→LLM flow, and Docker-friendly deployment scripts.
How we checked2 sources · 2/3 runs agreed · evidence score 56

How the score was reached

  • Partly verdict base52
  • 2 cited sources+1
  • Price verified on pricing page+3
  • Evidence score56

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 · 2

Every page the run actually retrieved.

Integrity checks

What held up, and what did not.

✓ Price read off the page! 2 of 3 runs agreed; the verdict is the majority✓ Citations limited to fetched pages! 1 moat recorded