NeuraLaunch
NeuraLaunch · Tabempa EngineeringDiscovery · OpenFrom lost to launched · 2026

You know something
needs to change. We’ll tell you what —
and walk it with you.

NeuraLaunch interviews your situation, commits to oneclear recommendation, and partners with you through every task — until you’ve shipped, learned, or decided what comes next.

Free to start · No card · ~12 minutes

Engine
Multi-cycle, multi-phase
Models
Opus · Sonnet · Haiku
Tools shipped
5 · Execute+
Recommendation count
One.
§ 01The cycle

One arc. From first question
to first outcome.

  1. 01

    Be heard.

    A 4-phase interview builds a fifteen-field belief state. Up to fifteen questions. Audience-aware. Capped, so it ends.

    DiscoveryOrientation → goals → constraints → conviction. Hard cap at 15.
  2. 02

    One recommendation.

    Synthesised across the full belief state. With reasoning, risks, assumptions you can flag, and the alternatives that were rejected — and why, for you specifically.

    Phase 1B emissionOpus 4.6 reasons, Sonnet formats. No menus. No hedge.
  3. 03

    Push back.

    Up to ten rounds of real argument. Defend, refine, replace. If you're right and the evidence agrees, the recommendation changes.

    Pushback engineTwo-phase. Hard cap triggers a constrained alternative.
  4. 04

    A real roadmap.

    Phased. Sequenced. Sized to the hours you actually have. Each task carries reasoning, success criteria, and the tools we'll use to get through it.

    Roadmap engineUp to five phases. Up to five tasks per phase.
  5. 05

    Learn. Continue.

    At the end of every cycle: what happened, what we got wrong, what the evidence says, the forks ahead, the parking lot. The next cycle starts smarter than the last.

    • “The recommendation read me back to myself.” — placeholder, backfill from stories archive.
    • “Got the day-job-friendly cycle I needed.” — placeholder, backfill from stories archive.
    • “Pushback round 6 changed everything.” — placeholder, backfill from stories archive.
    Continuation briefFive sections. Auto-generated. Ready to read.
§ 02A recommendation

This is what a real one
looks like.

Not a menu. One direction, for your specific situation, with reasoning, risks, and what would make it wrong — all on the table.

Recommendation · Cycle 01Confidence · High

Spend the next three weeks proving demand for the maternal-care advisory line. Not building the app.

You have a clinical network, a real WhatsApp following, and a shortage of hours. The app will eat 200 hours before anyone says yes. A three-week paid advisory pilot, run from your existing channel, will tell you within thirty days whether to build at all — and at what price.

First three steps

① Frame the pilot as a 30-day paid line.
② Recruit ten existing followers at NLe 200/mo.
③ Run consultations via WhatsApp Business.

Time to first result

14 days. First paid consultation booked.

Sized against the 8 weekly hours you stated.

What would make this wrong

If < 3 of 10 followers say yes at price, the advisory frame is wrong — pivot to free-tier triage with escalation, not a paid line.

3 assumptions · flaggable2 alternatives · rejectedPush back · 10 rounds
§ 03The toolkit

Five tools, for the work
that decides the outcome.

  • 001Conversation CoachOpusRehearse the pitch. Channel-native opening, anticipated objections, fallback positions — then a role-play in character.
  • 002Outreach ComposerSonnetWhatsApp, email, LinkedIn. Single, batch, or D1 / D5 / D14 sequence — each with a note on why it works.
  • 003Research ToolOpus · 25 stepPlain-language query, structured findings, source URLs, confidence labels — verified, likely, unverified.
  • 004Service PackagerSonnetThree priced tiers from your situation. Starter, Pro, Premium — with revenue scenarios and reasoning.
  • 005Validation PagePublicA live landing page with surveys and analytics. Real demand signal before you write a line of code.
§ 04Price

Each tier unlocks the next
layer of the journey.

Founding member rates

First 50 users — Execute at $19/month forever, Compound at $29/month forever. 46 of 50 slots remaining.

FreeTrust

Your first honest answer.

$0One interview. One recommendation. Yours to keep.
  • Two discovery interviews
  • One full recommendation with reasoning
  • Alternatives rejected, with why
  • Honest falsification — what would make this wrong
  • Inline assumption flagging
Start free
ExecuteMost chosen

From recommendation to revenue.

$279/year$23.25/mo equivalent · save $69 (20%)
  • Push back up to 10 rounds
  • Phased execution roadmap
  • Inline assumption flagging
  • The toolkitMonthly tool limits apply — see Fair Use below.
  • Conversation Coach
  • Outreach Composer
  • Research Tool
  • Service Packager
  • Validation Page
  • Task check-ins + diagnostic chat
  • Proactive nudges + push notifications
  • Parking lot for adjacent ideas
  • A second voice before you pause — three-mode reflection on why
  • Continuation brief at cycle end — drawn from your check-ins, outreach activity, research patterns, validation signal, and parking lot
  • Fork selection into the next cycle
  • Transformation report — a written narrative of what happened, ready to publish
  • 1 active venture (up to 2 paused)
CompoundMemory

The system gets smarter.

$470/year$39.17/mo equivalent · save $118 (20%)
  • Everything in Execute, plusBuilt for founders running more than one bet at a time.
  • Push back up to 15 rounds (vs 10)
  • 3 active ventures, up to 4 paused
  • Memory that learns across all your ventures
  • Voice mode in every text surface
  • Tool capacityFair use details below.
  • 3× the monthly tool quota of Execute
Fair use limits — what’s included in each tier.

  • Research Tool

    Execute
    30 / month
    Compound
    100 / month

  • Conversation Coach

    Execute
    50 / month
    Compound
    150 / month

  • Outreach Composer

    Execute
    100 / month
    Compound
    300 / month

  • Service Packager

    Execute
    20 / month
    Compound
    60 / month

  • Voice transcription

    Execute
    —
    Compound
    30 / hour

These ceilings exist so paid plans stay sustainable as LLM costs scale. Most founders use a fraction. If you’re consistently hitting them, upgrade or talk to us.

Eight weeks from today

You’ll know.

Either you’ve shipped. Or you’ve learned what to change. Or you’ve decided this isn’t the path — with proof. Whichever it is, you won’t be guessing anymore.

Begin Discovery