Skip to content
Sign in
SWC 2026 FinalistGrand Final: Lunch Block

Ketzai

Curriculum-aligned generative AI that already reaches 40,000+ teachers; the open question is who pays.

Santo Domingo, Dominican Republic, Latin AmericaEdTech, AI lesson planning for teachersDominican Republic – Santo Domingo 2026 (1st place)
Stage
Pre-seedAI at the core
Product
On the marketStrong traction
Funding raised
Undisclosedamount not disclosed
Founded
—
Founders
3Andy, José…
Global score
5.0Grade CRank 39 of 151

Overview

Ketzai, founded by Andy Caballero, is an AI platform that helps teachers plan their lessons and reduce their administrative workload.

Traction

Strong traction

More than 40,000 teachers use Ketzai Teacher Lab (according to the website), aligned with the official MEDUCA curriculum in Panama.

Market

AI in education

Teachers in Latin America spend unpaid hours producing lesson plans that must align with national curricula; ministries are under pressure to raise quality without adding staff. Generative AI makes curriculum-aligned planning cheap, but consumer teachers rarely pay, so scale revenue usually comes from ministries, school networks, publishers or development funders. Procurement is slow and public, but a single ministry contract can cover a whole country's teachers.

Addressable market

2025
$5.4B
2030
$9.6B
+12.2% a year

AI in education: global market estimated at $5.4bn in 2025 and $9.6bn in 2030 (360iResearch, CAGR 12.2%).

Customers

Public and private school teachers; school networks; ministries of education (MEDUCA Panama referenced); publishers

Who signs the cheque

Ministries and school networks (B2G/B2B); individual teachers on freemium

Geographies

  • Panama
  • Dominican Republic
  • Spanish-speaking Latin America

Drivers

  • Teacher workload and shortages
  • Ministry push to align teaching with official curricula
  • Cheap generative AI enabling per-country localisation

Headwinds

  • Monetisation of teachers is weak; public procurement cycles are long
  • Free tools from Google/Microsoft and global EdTech
  • Curriculum differences limit copy-paste expansion

Global AI in education market estimated at $5.4bn in 2025 and $9.6bn in 2030 (360iResearch, CAGR 12.2%); 2025 range of $5.4bn to $18.9bn depending on the research firm (Mordor $6.9bn).

Scalability 7/10. AI platform replicable across the Spanish-speaking world, but aligned with national curricula.

Market source: cn.gii.tw

Technology

Category

Vertical generative AI SaaS (EdTech)

AI at the coreGenerative AI / LLM

Maturity

  1. R&D
  2. Prototype
  3. Pilot
  4. On the market
  5. Scaling

in market

How it works

LLM generation (Google Gemini API) constrained by official curricula produces lesson plans, teaching sequences and guides. The asset is the structured curriculum mapping per country.

Moat

Moderate: the curriculum mapping, ministry relationships (SENACYT, Ciudad del Saber) and the teacher base are defensible locally; the generation layer is not.

Differentiation

Aligned with national curricula (MEDUCA) rather than generic prompts; strong grassroots teacher adoption.

Integration points

  • School LMS and ministry platforms
  • Assessment engines
  • Content/lab providers that can be embedded in lesson plans

Depends on

  • Google Gemini API
  • Access to updated official curricula
  • Ministry endorsement

Use of AI

Generative AI that produces lesson plans and teaching sequences aligned with official curricula.

Product in detail

Generative AI producing lesson plans, teaching sequences and guides aligned with official curricula (e.g. MEDUCA Panama).

Business

Model
Freemium to teachers; monetisation 'not specified' in the record, so institutional licences are the logical path.
Revenue logic
Per-teacher or per-school licence funded by ministry/network; unit cost driven by LLM tokens per plan.
Go-to-market
Bottom-up teacher adoption used as proof for top-down ministry and school-network deals; country-by-country curriculum packs.

Next milestones

  1. First paid ministry or school-network contract
  2. Second country curriculum pack beyond Panama
  3. Seed round

Key risks

  • Unclear monetization
  • No capital
  • Dependency on a single LLM provider for cost and quality

Opportunities

5 found in the network

Concrete moves for Ketzai inside the network: who to sell to, sell through, build with or raise from, each with the reasoning, the evidence and the first step.

  • Customers 1
  • Distribution 1
  • Co-development 1
  • Investors 2

Customers1

Who could buy, or sell through

Control LLM unit costs before scaling free usage

With free teacher usage on a paid Gemini plan, Ketzai's cost per plan determines whether ministry pricing is viable. UnitSense attributes and routes spend, giving Ketzai a measured cost base for contracts.

30%confidence

Evidence from the records

  • “Google Gemini API (paid plan), declared as AI provider in the privacy policy”
  • “claims 35% average savings through smart routing on real traffic”
If it works
A clear cost per teacher per month to anchor B2G pricing.
First step
Run a one-week cost audit of Ketzai's generation traffic through the UnitSense gateway.

Distribution1

A channel or customer base to reach together

Embed virtual labs into 40,000 teachers' lesson plans

PraxiLabs sells 3D virtual science labs globally; Ketzai generates science lesson plans for teachers who often lack physical labs. Auto-inserting PraxiLabs experiments into generated plans creates a distribution channel for PraxiLabs and a revenue-share stream for Ketzai.

35%confidence

Evidence from the records

  • “3D virtual science labs”
  • “More than 40,000 teachers use Ketzai Teacher Lab”
If it works
Referral/revenue share per school activated; Ketzai gains its first monetised feature.
First step
Propose a revenue-share pilot with PraxiLabs on Panama's science curriculum units.

Co-development1

An integration or joint product

Plan-assess loop: Ketzai planning + Maria assessment

Ketzai produces lesson plans; Maria Educação diagnoses student performance and builds learning paths for publishers and municipalities. Linking assessment results to lesson planning closes the loop that ministries buy, and gives Maria a Spanish-speaking entry route through Ketzai's teacher base.

40%confidence

Evidence from the records

  • “Generative AI producing lesson plans, teaching sequences and guides aligned with official curricula (e.g. MEDUCA Panama).”
  • “São Paulo EdTech that uses AI to assess and diagnose student performance and then build adapted learning paths, from basic to higher education.”
If it works
A joint ministry-grade offer (plan, teach, assess) that is easier to sell to governments than two point tools.
First step
Map Ketzai's plan schema to Maria's assessment outputs and draft a one-page joint offer for a Panamanian pilot.

Investors2

A fund whose thesis fits

Central American EdTech pre-seed

Innogen invests pre-seed/seed in EdTech across Central America and Latin America; Ketzai's Panama footprint fits its geography. Innogen backed network member Pulster.ai (Guatemala), which offers a warm introduction.

30%confidence

Evidence from the records

  • “aligned with the official MEDUCA curriculum in Panama”
  • “Broad teacher adoption”
If it works
A first equity round to fund monetisation and a second country.
First step
Ask Pulster.ai's founders for an intro to Innogen Capital with a ministry-pipeline one-pager.

Emerging-market EdTech fit

Seedstars invests very early in scalable emerging-market models reaching many people, explicitly including EdTech and Latin America, with tickets of $0.15–0.25M. It backed Baniql in the network.

25%confidence

Evidence from the records

  • “Invest very early in emerging-market tech ecosystems, in scalable models reaching many people.”
  • “Interesting adoption but model still to be proven.”
If it works
Small first cheque plus emerging-market playbook for B2G sales.
First step
Submit Ketzai to Seedstars with teacher-usage cohort data and a monetisation plan.

Analyst reading of 21 Nov 2026, from the public record of the network. Method: analyst-ai.

Investors that fit

Seedstars' Geneva seed fund ($20M first close) investing $150K-$250K very early in scalable emerging-market startups.

AI lesson planning used by 40,000+ teachers, aligned with Panama's curriculum: emerging-market edtech at scale.

40%confidence

Evidence from the records

  • “More than 40,000 teachers use Ketzai Teacher Lab”
  • “AI lesson planning for teachers”

CORFO's equity-free Santiago accelerator (since 2010) supporting early-stage founders from 85 countries through Build, Ignite and Growth programmes.

AI lesson-planning used by 40,000+ teachers in Panama: LatAm edtech with traction, a Chile expansion candidate.

35%confidence

Evidence from the records

  • “AI lesson planning for teachers”
  • “Pre-seed”
First step
Apply to Ignite with a Chilean school pilot.

San Salvador VC whose $11M IDB Lab-backed fund backs ~25 early-stage startups in El Salvador, Honduras and Guatemala, ≥25% women-led.

Dominican AI lesson-planning for teachers — edtech, but outside the three target countries.

30%confidence

Evidence from the records

  • “"EdTech"”
  • “AI lesson planning for teachers”
  • “More than 40,000 teachers use Ketzai Teacher Lab (according to the website), aligned with the official MEDUCA curriculum in Panama.”
Intro path
Founder-to-founder intro via Pulster.ai (network portfolio company of Innogen Capital Ventures).
First step
Only if expansion to Guatemala/El Salvador is planned.

Egypt's first edtech corporate VC (Nahdet Misr with ASRT) offering incubation, acceleration and follow-on investment; backed Praxilabs' Series A.

AI lesson planning used by 40,000+ teachers, curriculum-aligned in Panama: EdTech with curriculum alignment Nahdet Misr understands; needs Arabic/Egypt localisation.

20%confidence

Evidence from the records

  • “AI lesson planning for teachers”
  • “More than 40,000 teachers”
Intro path
Praxilabs (EdVentures portfolio)
First step
Explore an Egyptian curriculum pilot via Praxilabs.

Colombian family compensation fund whose venture arm, Comfama Ventures, led the equity in Monet's US$24M Series A.

AI lesson planning used by 40,000+ teachers: education services, Dominican Republic/Panama.

15%confidence

Evidence from the records

  • “AI lesson planning for teachers”
  • “More than 40,000 teachers”

Investment analysis

5.0C

Grade C

Rank 39 of 151

Disruption4.0/10
Business model5.0/10
Market6.0/10
Risk(lower is better)6.0/10

Global score = mean of disruption, model, market and (11 − risk). Last updated 21 Nov 2026.

Interesting adoption but model still to be proven.

Disruption

4/10

AI lesson planning: known category.

Business model

5/10

More than 40,000 teachers claimed, monetization not specified.

Market

6/10

TAM AI in education ~$5.37bn (base 5), medium scalability 7/10 (0), international expansion: used in Panama (MEDUCA curriculum) from the Dominican Republic (+1) → 6/10.

Risk

6/10

Lower is better

Medium-high risk: no funding round, uncertain monetization.

Strengths

  • Broad teacher adoption
  • Alignment with official curricula

Risks

  • Unclear monetization
  • No capital

Funding

Undisclosed, amount not disclosed
  1. 2025

    grant/prize (not quantified)not disclosed

    SENACYT (“Ganadores I+D 2025”, according to the website)

  2. non daté

    incubation / accelerationnot disclosed

    Ciudad del Saber, Emprelatam, Google for Startups

  3. 2026

    prizenon-monetary (qualification for the US$1M final)

    Startup World Cup Dominican Republic

Amount used in the study. SENACYT grant claimed without amount; no equity funding round.

Note. no public funding round found (backing not quantified)

SWC Dominican Republic 2026 win (Hotel Jaragua, Santo Domingo) confirmed. Note: the website targets Panama exclusively (MEDUCA, 40,000+ teachers, Ciudad del Saber, SENACYT, CAPATEC) – the headquarters may be in Panama even though the press calls it Dominican; country left unchanged for lack of direct evidence.

Team

  • Andy Caballero

    Founder & CEO

    Systems and computer engineer specializing in AI.

  • José Andrés Aragón

    COO

    Industrial engineer (strategy, finance, operations).

  • Helen Guerrero

    CTO

    Systems and computer engineer, software development.

Sources

Regional final
Dominican Republic – Santo Domingo 2026 (1st place)
Identification confidence
high
  1. 1ecosistemastartup.com/ketzai-ia-educativa-dominicana-gana-el-startup-world-cup/
  2. 2ketzai.com/

From the open study Startup World Cup 2026 by TIBOK, based on public sources, verified in November 2026. Scores are an analyst's judgment on an explicit grid, not the jury's.

Wall

Wall

No posts yet