Developer docs
API playgroundTry for free, no card

Search company profiles

Eedi

Full company profile

uuid0000sih

Namestring
Eedi
Legal namestring
Eedi Ltd
Websiteurl
eedi.com
Company typeenum
Private
Founded yearint
2015
Descriptiontext

Eedi Labs is a UK-based applied research organization founded in 2015 that builds AI-powered diagnostic infrastructure for mathematics education. Its core asset is a diagnostic engine refined over nearly a decade of classroom use across more than one billion learner interactions, comprising a library of over 60,000 multiple-choice diagnostic questions where each incorrect distractor is mapped to a specific named misconception. This engine is paired with proprietary Knowledge Tracing (KT) models — including a compact 0.73M-parameter LLM KT that achieves 72.8% accuracy on student response prediction versus 58.6% for GPT-4o-mini — enabling precise, real-time identification of what each student misunderstands and why.

The company operates through a tiered product surface: Eedi School (the free testbed platform for classrooms, homework, and RCTs); the DQR (Diagnostic Question Response) app, which delivers whole-class formative assessment using only a teacher's device; the Constrained AI Tutor, a human-in-the-loop AI tutoring system that activates only when a student answers a diagnostic check-in incorrectly and is bounded to the specific construct being learned; and Eedi Inside, an OEM/white-label program that embeds the diagnostic engine inside third-party publisher and AI platforms. Eedi School Plus, a £54/student/year subscription for parents and students, adds chat-based human tutor support.

Eedi monetizes through a hybrid model combining freemium access for schools and teachers, direct parent/student subscriptions, and licensing of the diagnostic engine to publishers (notably Imagine Learning's Imagine IM US math curriculum, back-to-school 2026) and frontier AI labs (Google DeepMind's LearnLM research). It primarily serves K-12 secondary schools (UK Years 7-10, US grades 6-8), with a stated mission to improve learning outcomes for one billion children by 2030. The company employs approximately 40 people on a fully remote UK basis and is currently expanding into the US market via the Imagine Learning distribution deal and an Accelerate-funded RCT.

Short descriptiontext

Eedi Labs builds an AI-powered diagnostic engine for K-12 mathematics that identifies specific student misconceptions in real time using over 60,000 mapped questions and proprietary Knowledge Tracing models. The company serves UK and US secondary schools directly, parents via subscriptions, and publishers through OEM licensing.

Operating statusenum
Operating
Ownership categoryenum
Headcount rangeband
51–100
akta.pro rankint
HeadquartersLondon, United Kingdom
HQ citystring
London
HQ countrystring
United Kingdom
HQ regionstring
Europe
Markets served

Serves global market

Offices2 records

Each record includes

City, Country, Type, Description, Source

Keyword5 values
AI tutoring platform, math diagnostic engine, knowledge tracing models, misconception identification, adaptive learning assessment
Industry1 code
PRIMARYEDAFACAF
Digital Labs, Simulations & Virtual Practicals (Non-XR)
Education/Education Technology (EdTech) & Digital Learning/Digital Curriculum, Courseware & Interactive Learning Content
NAICS code2 codes
  • Educational Services611
  • Educational Support Services611710
SIC code2 codes
  • Services-Educational Services8200
  • Services-Computer Integrated Systems Design7373
Product category
EdTech AI Infrastructure
Social media profiles1 record
GTM motion4 records

Each record includes

Type, Description, Source

Revenue model4 records
1School/Teacher Subscriptions (Free Tier)
TypeFreemium
Description

Eedi School platform is provided free to schools and teachers. Schools and teachers can access the platform at no cost to use diagnostic quizzes, assign homework, and access AI tutoring support.

eedi.com
2Eedi School Plus (Parent/Student Subscription)
TypeSubscription Recurring
Description

Parents and students can purchase paid monthly subscriptions directly. Eedi School Plus gives students unlimited access to chat-based human tutor support (9 AM – 9:30 PM UK time). Cost is £54 per student per year.

eedi.com
3Intelligence Infrastructure / Licensing
TypeLicensing Royalties
Description

Eedi embeds its diagnostic engine as the intelligence layer inside third-party platforms — publishers, LLMs, and learning platforms. This is an OEM/white-label licensing model where Eedi's technology powers partner products.

eedi.com
4Partnership / Integration Revenue
TypeLicensing Royalties
Description

Eedi's partnership with Imagine Learning integrates real-time AI assessments into the Imagine IM math curriculum. Revenue derives from licensing the diagnostic engine to Imagine Learning for embedding in their core curriculum product.

prnewswire.com
Marketing channels6 records

Each record includes

Title, Type, Stage, Description, Source

Distribution channels5 records

Each record includes

Title, Type, Scope, Target buyer, Description, Source

Cost components5 values
Personnel, Technology or R&D, Infrastructure, Marketing or Sales, Operations
Pricing details3 tiers
1Free for schools and teachers
ModelFreemiumBilling cadenceOthers
Notes

No charge to schools or teachers. Platform access including diagnostic quizzes, assignments, and reporting for classroom use.

eedi.com
2Eedi School Plus — Parent/Student subscription
ModelSubscriptionBilling cadenceAnnual
Notes

Eedi School Plus subscription at £54 per student per year. Includes unlimited access to chat-based human tutor support, 9 AM – 9:30 PM UK time. Supports 1 quiz per week (120+ questions annually) target for engaged use.

eedi.com
37-day free trial for parent/student subscriptions
ModelFreemiumBilling cadenceOthers
Notes

A free 7-day trial period is offered to give parents/students the opportunity to assess the suitability of the service before committing to a paid subscription.

GTM typeB2B
B2B
Offering typeSoftware
Software
Brand1 record
1Eedi School
Description

Testbed platform and classroom/homework tool powered by Eedi's diagnostic engine for real-time student misconception identification and AI tutoring.

eedi.com
Core offering1 text field

Eedi builds and operates an AI-powered diagnostic engine for mathematics that identifies and resolves individual student misconceptions in real time using a library of 60,000+ mapped questions and proprietary Knowledge Tracing models. The company delivers this through three product lines: Eedi School (free classroom and homework platform for teachers, students, and parents), Eedi Inside (OEM/embedded licensing of the diagnostic engine for publishers, frontier model providers, and learning platforms), and the Constrained AI Tutor (human-in-the-loop tutoring that activates when diagnostic checks are failed).

Differentiator
Functional benefit
Problem solved
Quantifiable outcome1 of 9 values shown
  • Effect size d = 0.34 (CACE) for actively engaged students — equivalent to 113% of a typical student's annual maths growth
+8 more records
Product overview1 text field

Eedi is an applied research organization that operates primarily through its Eedi School platform and the Eedi Diagnostic Engine as its core infrastructure. The diagnostic engine, built on over a decade of data and one billion learner interactions, identifies student misconceptions in real time through a library of 60,000+ questions. Eedi Inside extends this intelligence to third-party publishers and platforms as infrastructure. The Constrained AI Tutor module works within Eedi School to provide targeted intervention when misconceptions are detected. The DQR system enables whole-class diagnostic assessment through teacher devices. The company offers BUILD (infrastructure for partners), TEACH (DQR for teachers), and LEARN (research and tutoring) product lines.

Product and service5 records
1Eedi School
CategoryEdTech / Educational Software Platform
Description

A free-to-schools classroom and homework platform powered by Eedi's diagnostic engine, used by teachers to assign diagnostic quizzes, by students to complete them, and by parents to monitor progress. Available to UK and US K-12 schools at no charge.

2Eedi Inside (Diagnostic Engine Intelligence Infrastructure)
CategoryEdTech AI Infrastructure / Licensing
Description

An OEM/white-label licensing program that embeds Eedi's diagnostic engine as the intelligence layer inside third-party publisher platforms, LLM-powered products, and learning systems. The engine drives efficacy for LLMs, assessment tools, and AI tutoring systems.

3Eedi School Plus (Parent/Student Subscription)
CategoryEdTech / Educational Subscription
Description

Paid annual subscription for parents and students priced at £54 per student per year. Includes unlimited access to chat-based human tutor support and additional AI tutoring features beyond the free school-tier offering.

4Constrained AI Tutor
CategoryEdTech / AI Tutoring Software
Description

An AI-powered tutoring tool that activates only when a student answers a diagnostic check-in question incorrectly and engages the student in brief text-based conversations to resolve specific mathematical misconceptions before returning them to the lesson. Features human-in-the-loop supervision where human tutors review, edit, or reject every AI-drafted message.

5DQR (Diagnostic Question Recommendation) Classroom Tool
CategoryEdTech / Classroom Assessment Software
Description

A classroom diagnostic tool that enables teachers to gather real-time formative assessment data from an entire class without requiring students to have their own devices. Designed for low-device classrooms and offline-first use cases.

Scale indicator13 records

Each record includes

Type, Value, Description, Source

Partnership7 partners
Strategic tierCoreTypeStrategic or Co-development PartnerAnnounced on2026-05-13
Description

Eedi was selected as a grantee of Accelerate's 2026-27 Call for Effective Technology (CET) program. Selected from 87 applicants, Eedi will conduct its first US RCT of its constrained AI Tutor in middle school classrooms in two US school districts. The program is designed to generate evidence meeting ESSA Tier 1 certification requirements. This represents Eedi's formal expansion into the US K-12 market.

Strategic tierCoreTypeOEM/ Whitelabel/ Licensing PartnerAnnounced on2026-01-15
Description

Imagine Learning and Eedi announced a strategic partnership to integrate Eedi's AI-powered diagnostic engine directly into Imagine Learning's Imagine IM math curriculum for US classrooms. The system launches for back-to-school 2026. Eedi's engine uses over a decade of student learning data to identify student misconceptions in as few as 10 questions, enabling teachers to adjust instruction immediately without additional testing tools. The partnership aims to expand real-time assessment across Imagine Learning's portfolio including ELA and science.

Strategic tierCoreTypeStrategic or Co-development PartnerAnnounced on2025-11-11
Description

Eedi partnered with Google DeepMind to conduct exploratory RCTs on human-in-the-loop AI tutoring in UK classrooms. The research found that supervised AI tutoring matches the effectiveness of human tutors. Eedi's diagnostic engine provides the specific misconception signal that turns a generic LLM (LearnLM) into a pedagogically-grounded AI tutor. Supervising tutors approved 74.4% of AI-drafted messages without edits. A second RCT was launched in April 2026 with DeepMind, involving 1,525 students across 10 UK secondary schools.

4WhatWorked Education
Strategic tierCoreTypeStrategic or Co-development PartnerAnnounced on2025-11-07
Description

WhatWorked Education, led by Dr. Wayne Harrison and Professor Steve Higgins, conducted the independent gold-standard RCT evaluating Eedi's impact on 2,901 students across 20 UK schools. This research validated Eedi's effectiveness and formed the basis of the EduEvidence gold efficacy certificate.

eedi.com
Strategic tierNotableTypeStrategic or Co-development Partner
Description

Stanford University's SCALE Initiative reviewed over 800 papers on AI in K-12 education and included Eedi's 2025 AI tutor RCT as one of only 20 high-quality causal studies. Stanford's publishing of research notes about the Eedi/Google DeepMind study also increased Eedi's visibility in the academic community.

Strategic tierMinorTypeStrategic or Co-development Partner
Description

Rising Academies is an innovative education provider operating in Rwanda. Eedi's Simon Woodhead visited Rising Academies' schools during the Global Learning Conference 2026 in Kigali and observed their use of AI tutors (RORI) on mobile phones. Eedi is developing an offline-first version of its DQR app to address data cost concerns raised by Rwandan teachers, working with lessons learned from Rising Academies' experience.

7van der Schaar Lab (University of Cambridge)
Strategic tierNotableTypeTechnology or Integration
Description

Mihaela van der Schaar's lab at Cambridge is a leading AI for education research lab. Eedi's collaboration with the lab advances AI for human empowerment, making learning more personal and accessible. The partnership contributes to Eedi's research credibility and open science commitment.

eedi.com
Recent move6 records

Each record includes

Date, Type, Title, Description, Source

Expansion highlight6 records

Each record includes

Type, Description

Peers10 records
TypeDirect peer
Description

Major US K-12 curriculum publisher and partner of Eedi; integrates Eedi's diagnostic engine into the Imagine IM math curriculum for back-to-school 2026. Comparable as a K-12 math content and assessment provider serving US school districts.

TypeDirect peer
Description

Free K-12 learning platform with AI tutoring capabilities (Khanmigo) and broad subject coverage. Directly competes with Eedi School on free math practice and AI tutoring in US/UK K-12 markets, with comparable freemium GTM motion.

TypeDirect peer
Description

Free K-6 math learning platform aligned to US curricula, used widely in US school districts. Directly comparable to Eedi's free-to-schools model and focus on targeted math skill-building for low-attaining students.

TypeDirect peer
Description

Subscription-based adaptive K-12 math and ELA practice platform with diagnostic question design. Comparable to Eedi on diagnostic question libraries, adaptive learning, and district-level sales motion, though broader in subject coverage.

TypeDirect peer
Description

Adaptive K-8 math platform using intelligent adaptive learning to identify and address student misconceptions. Direct competitor in diagnostic-driven math instruction, with shared focus on misconception-based pedagogy and US school district GTM.

TypeDirect peer
Description

AI-driven math tutoring (MATHia) for middle and high school, with adaptive misconception diagnosis and proven RCT efficacy. Closely comparable to Eedi on math-specific AI tutoring, evidence-based claims, and US secondary school GTM.

TypeDirect peer
Description

Adaptive math learning platform with step-by-step diagnostic feedback and AI tutoring for K-12. Direct comparable to Eedi on math-only focus, adaptive diagnostic questioning, and emphasis on misconception remediation.

TypeDirect peer
Description

Adaptive AI tutoring platform originating in China using Bayesian knowledge tracing and diagnostic engines for K-12 math. Comparable to Eedi on KT-driven adaptive learning, though operates primarily in Asian markets.

TypeBroad incumbent
Description

Large-scale consumer and school study platform with adaptive practice and assessment features across many subjects. Overlaps with Eedi on practice-question-driven learning and freemium GTM, though broader in subject scope.

TypeEmerging player
Description

EdTech operator in Rwanda using RORI, an AI tutor on mobile phones, with whom Eedi is co-developing an offline-first DQR app. Comparable as a mission-driven AI tutoring deployment in cost-sensitive emerging markets.

Market position
Strengths5 records

Each record includes

Headline, Details, Source

Weaknesses5 records

Each record includes

Headline, Details, Source

Competitive moat5 records

Each record includes

Type, Details

Key risks7 records

Each record includes

Headline, Details, Source

Key highlights7 records

Each record includes

Headline, Details, Source

Customer concentration

Classification, Details

Named customers7 records

Each record includes

Name, Industry, Type, Use case, Source, UUID

Segment5 records

Each record includes

Title, Type, Primary, Description, Pain point addressed, Use case, Source

Ideal customer profile4 records

Each record includes

Profile, Firmographic size, Sales motion, Sales cycle length, Buying structure, Purchase trigger, Buyer persona, Geography, Industry vertical, Primary use case, Description, Pain points, Evidence proof points, Target buyer

Technology focused
Yes
API detail
Has APIbool
No

Docs URL, Description

Integration2 records

Each record includes

Title, Type, Description, Source

AI capability6 records

Each record includes

Type, Description, Source

AI maturity
App detail

Has app

Feature6 records

Each record includes

Title, Differentiator, Description, Source

Core technology
Revenue estimate
Valuation estimate
Number of profiles
Profiles4 records

Each record includes

Name, Designation, Designation category, Overview, Profile commentary, Source

No data
Compliance1 record

Each record includes

Name, Class, Description

Funding overview

Funding stage, Last funding date, Total funding USD

Funding rounds3 records

Each record includes

Round, Amount USD, Date, Pre money valuation, Total investors, Investors, News

Investors5 records

Each record includes

Name, Type, Date of entry, Rounds participated, Website

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Each record includes

Name, Acquisition type, Announced date, Completed date, Status, Website, News

Investment

Each record includes

Name, Round, Announced date, Lead investor, Website, News

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Eedi

EdTech AI Infrastructureeedi.com

Eedi Labs builds an AI-powered diagnostic engine for K-12 mathematics that identifies specific student misconceptions in real time using over 60,000 mapped questions and proprietary Knowledge Tracing models. The company serves UK and US secondary schools directly, parents via subscriptions, and publishers through OEM licensing.

What Eedi does

Eedi Labs is a UK-based applied research organization founded in 2015 that builds AI-powered diagnostic infrastructure for mathematics education. Its core asset is a diagnostic engine refined over nearly a decade of classroom use across more than one billion learner interactions, comprising a library of over 60,000 multiple-choice diagnostic questions where each incorrect distractor is mapped to a specific named misconception. This engine is paired with proprietary Knowledge Tracing (KT) models — including a compact 0.73M-parameter LLM KT that achieves 72.8% accuracy on student response prediction versus 58.6% for GPT-4o-mini — enabling precise, real-time identification of what each student misunderstands and why.

The company operates through a tiered product surface: Eedi School (the free testbed platform for classrooms, homework, and RCTs); the DQR (Diagnostic Question Response) app, which delivers whole-class formative assessment using only a teacher's device; the Constrained AI Tutor, a human-in-the-loop AI tutoring system that activates only when a student answers a diagnostic check-in incorrectly and is bounded to the specific construct being learned; and Eedi Inside, an OEM/white-label program that embeds the diagnostic engine inside third-party publisher and AI platforms. Eedi School Plus, a £54/student/year subscription for parents and students, adds chat-based human tutor support.

Eedi monetizes through a hybrid model combining freemium access for schools and teachers, direct parent/student subscriptions, and licensing of the diagnostic engine to publishers (notably Imagine Learning's Imagine IM US math curriculum, back-to-school 2026) and frontier AI labs (Google DeepMind's LearnLM research). It primarily serves K-12 secondary schools (UK Years 7-10, US grades 6-8), with a stated mission to improve learning outcomes for one billion children by 2030. The company employs approximately 40 people on a fully remote UK basis and is currently expanding into the US market via the Imagine Learning distribution deal and an Accelerate-funded RCT.

Eedi firmographics

Firmographics
Name
Eedi
Legal name
Eedi Ltd
Website
https://eedi.com
Company type
Private
Founded year
2015
Operating status
Operating
Headcount range
51–100 employees
Short description
Eedi Labs builds an AI-powered diagnostic engine for K-12 mathematics that identifies specific student misconceptions in real time using over 60,000 mapped questions and proprietary Knowledge Tracing models. The company serves UK and US secondary schools directly, parents via subscriptions, and publishers through OEM licensing.
Ownership category
akta.pro rank

Eedi industry classification

Industry
Product category
EdTech AI Infrastructure
NAICS
Educational Services (611), Educational Support Services (611710)
SIC
Services-Educational Services (8200), Services-Computer Integrated Systems Design (7373)
akta.pro industry
PRIMARYEDAFACAF
Digital Labs, Simulations & Virtual Practicals (Non-XR)
Education/Education Technology (EdTech) & Digital Learning/Digital Curriculum, Courseware & Interactive Learning Content

Keywords

  • AI tutoring platform
  • Math diagnostic engine
  • Knowledge tracing models
  • Misconception identification
  • Adaptive learning assessment

Where Eedi is headquartered

Location

Headquarters

HQ city
London
HQ country
United Kingdom
HQ region
Europe

Offices2 records

Markets served

Eedi business model

Business model
GTM type
B2B
Offering type
Software
Cost components
Personnel, Technology or R&D, Infrastructure, Marketing or Sales, Operations

Revenue model

  1. School/Teacher Subscriptions (Free Tier): Eedi School platform is provided free to schools and teachers. Schools and teachers can access the platform at no cost to use diagnostic quizzes, assign homework, and access AI tutoring support.
  2. Eedi School Plus (Parent/Student Subscription): Parents and students can purchase paid monthly subscriptions directly. Eedi School Plus gives students unlimited access to chat-based human tutor support (9 AM – 9:30 PM UK time). Cost is £54 per student per year.
  3. Intelligence Infrastructure / Licensing: Eedi embeds its diagnostic engine as the intelligence layer inside third-party platforms — publishers, LLMs, and learning platforms. This is an OEM/white-label licensing model where Eedi's technology powers partner products.
  4. Partnership / Integration Revenue: Eedi's partnership with Imagine Learning integrates real-time AI assessments into the Imagine IM math curriculum. Revenue derives from licensing the diagnostic engine to Imagine Learning for embedding in their core curriculum product.

Pricing tiers

ModelBillingPrice
FreemiumOthersFree for schools and teachers
SubscriptionAnnualEedi School Plus — Parent/Student subscription
FreemiumOthers7-day free trial for parent/student subscriptions

Go-to-market motion4 records

Distribution channels5 records

Marketing channels6 records

Eedi product offering

Product offering

Core offering

Eedi builds and operates an AI-powered diagnostic engine for mathematics that identifies and resolves individual student misconceptions in real time using a library of 60,000+ mapped questions and proprietary Knowledge Tracing models. The company delivers this through three product lines: Eedi School (free classroom and homework platform for teachers, students, and parents), Eedi Inside (OEM/embedded licensing of the diagnostic engine for publishers, frontier model providers, and learning platforms), and the Constrained AI Tutor (human-in-the-loop tutoring that activates when diagnostic checks are failed).

Product overview

Eedi is an applied research organization that operates primarily through its Eedi School platform and the Eedi Diagnostic Engine as its core infrastructure. The diagnostic engine, built on over a decade of data and one billion learner interactions, identifies student misconceptions in real time through a library of 60,000+ questions. Eedi Inside extends this intelligence to third-party publishers and platforms as infrastructure. The Constrained AI Tutor module works within Eedi School to provide targeted intervention when misconceptions are detected. The DQR system enables whole-class diagnostic assessment through teacher devices. The company offers BUILD (infrastructure for partners), TEACH (DQR for teachers), and LEARN (research and tutoring) product lines.

Differentiator

Problem solved

Functional benefit

Brands

  • Eedi School: Testbed platform and classroom/homework tool powered by Eedi's diagnostic engine for real-time student misconception identification and AI tutoring.

Products and services

  • Eedi School A free-to-schools classroom and homework platform powered by Eedi's diagnostic engine, used by teachers to assign diagnostic quizzes, by students to complete them, and by parents to monitor progress. Available to UK and US K-12 schools at no charge.
  • Eedi Inside (Diagnostic Engine Intelligence Infrastructure) An OEM/white-label licensing program that embeds Eedi's diagnostic engine as the intelligence layer inside third-party publisher platforms, LLM-powered products, and learning systems. The engine drives efficacy for LLMs, assessment tools, and AI tutoring systems.
  • Eedi School Plus (Parent/Student Subscription) Paid annual subscription for parents and students priced at £54 per student per year. Includes unlimited access to chat-based human tutor support and additional AI tutoring features beyond the free school-tier offering.
  • Constrained AI Tutor An AI-powered tutoring tool that activates only when a student answers a diagnostic check-in question incorrectly and engages the student in brief text-based conversations to resolve specific mathematical misconceptions before returning them to the lesson. Features human-in-the-loop supervision where human tutors review, edit, or reject every AI-drafted message.
  • DQR (Diagnostic Question Recommendation) Classroom Tool A classroom diagnostic tool that enables teachers to gather real-time formative assessment data from an entire class without requiring students to have their own devices. Designed for low-device classrooms and offline-first use cases.

Quantifiable outcome

  • Effect size d = 0.34 (CACE) for actively engaged students — equivalent to 113% of a typical student's annual maths growth
  • +8 more outcomes

Companies that use Eedi

Customer profile

Named customers7 records

Segments5 records

Ideal customer profiles4 records

Eedi technology and API

Technology

Technology focussed Yes

API detail

Has API
No
API docs
API detail

Core technology

AI maturity

App detail

Integration2 records

AI capability6 records

Feature6 records

Eedi partnerships and signals

Strategic signal

Partnerships

Seven partnerships are on record, tiered core, notable and minor.

  • AcceleratecoreStrategic or Co-development Partner · 13 May 2026Eedi was selected as a grantee of Accelerate's 2026-27 Call for Effective Technology (CET) program. Selected from 87 applicants, Eedi will conduct its first US RCT of its constrained AI Tutor in middle school classrooms in two US school districts. The program is designed to generate evidence meeting ESSA Tier 1 certification requirements. This represents Eedi's formal expansion into the US K-12 market.
  • Imagine LearningcoreOEM/ Whitelabel/ Licensing Partner · 15 January 2026Imagine Learning and Eedi announced a strategic partnership to integrate Eedi's AI-powered diagnostic engine directly into Imagine Learning's Imagine IM math curriculum for US classrooms. The system launches for back-to-school 2026. Eedi's engine uses over a decade of student learning data to identify student misconceptions in as few as 10 questions, enabling teachers to adjust instruction immediately without additional testing tools. The partnership aims to expand real-time assessment across Imagine Learning's portfolio including ELA and science.
  • Google DeepMindcoreStrategic or Co-development Partner · 11 November 2025Eedi partnered with Google DeepMind to conduct exploratory RCTs on human-in-the-loop AI tutoring in UK classrooms. The research found that supervised AI tutoring matches the effectiveness of human tutors. Eedi's diagnostic engine provides the specific misconception signal that turns a generic LLM (LearnLM) into a pedagogically-grounded AI tutor. Supervising tutors approved 74.4% of AI-drafted messages without edits. A second RCT was launched in April 2026 with DeepMind, involving 1,525 students across 10 UK secondary schools.
  • WhatWorked EducationcoreStrategic or Co-development Partner · 7 November 2025WhatWorked Education, led by Dr. Wayne Harrison and Professor Steve Higgins, conducted the independent gold-standard RCT evaluating Eedi's impact on 2,901 students across 20 UK schools. This research validated Eedi's effectiveness and formed the basis of the EduEvidence gold efficacy certificate.
  • Stanford University (SCALE Initiative)notableStrategic or Co-development PartnerStanford University's SCALE Initiative reviewed over 800 papers on AI in K-12 education and included Eedi's 2025 AI tutor RCT as one of only 20 high-quality causal studies. Stanford's publishing of research notes about the Eedi/Google DeepMind study also increased Eedi's visibility in the academic community.
  • Rising AcademiesminorStrategic or Co-development PartnerRising Academies is an innovative education provider operating in Rwanda. Eedi's Simon Woodhead visited Rising Academies' schools during the Global Learning Conference 2026 in Kigali and observed their use of AI tutors (RORI) on mobile phones. Eedi is developing an offline-first version of its DQR app to address data cost concerns raised by Rwandan teachers, working with lessons learned from Rising Academies' experience.
  • van der Schaar Lab (University of Cambridge)notableTechnology or IntegrationMihaela van der Schaar's lab at Cambridge is a leading AI for education research lab. Eedi's collaboration with the lab advances AI for human empowerment, making learning more personal and accessible. The partnership contributes to Eedi's research credibility and open science commitment.

Scale indicators13 records

Recent moves6 records

Expansion highlights6 records

Eedi competitors and assessment

Company assessment

Direct peers

  • Imagine Learning: Major US K-12 curriculum publisher and partner of Eedi; integrates Eedi's diagnostic engine into the Imagine IM math curriculum for back-to-school 2026. Comparable as a K-12 math content and assessment provider serving US school districts.
  • Khan Academy: Free K-12 learning platform with AI tutoring capabilities (Khanmigo) and broad subject coverage. Directly competes with Eedi School on free math practice and AI tutoring in US/UK K-12 markets, with comparable freemium GTM motion.
  • Zearn: Free K-6 math learning platform aligned to US curricula, used widely in US school districts. Directly comparable to Eedi's free-to-schools model and focus on targeted math skill-building for low-attaining students.
  • IXL Learning: Subscription-based adaptive K-12 math and ELA practice platform with diagnostic question design. Comparable to Eedi on diagnostic question libraries, adaptive learning, and district-level sales motion, though broader in subject coverage.
  • DreamBox Learning: Adaptive K-8 math platform using intelligent adaptive learning to identify and address student misconceptions. Direct competitor in diagnostic-driven math instruction, with shared focus on misconception-based pedagogy and US school district GTM.
  • Carnegie Learning: AI-driven math tutoring (MATHia) for middle and high school, with adaptive misconception diagnosis and proven RCT efficacy. Closely comparable to Eedi on math-specific AI tutoring, evidence-based claims, and US secondary school GTM.
  • Mathspace: Adaptive math learning platform with step-by-step diagnostic feedback and AI tutoring for K-12. Direct comparable to Eedi on math-only focus, adaptive diagnostic questioning, and emphasis on misconception remediation.
  • Squirrel AI: Adaptive AI tutoring platform originating in China using Bayesian knowledge tracing and diagnostic engines for K-12 math. Comparable to Eedi on KT-driven adaptive learning, though operates primarily in Asian markets.

Broad incumbents

  • Quizlet: Large-scale consumer and school study platform with adaptive practice and assessment features across many subjects. Overlaps with Eedi on practice-question-driven learning and freemium GTM, though broader in subject scope.

Emerging players

  • Rising Academies: EdTech operator in Rwanda using RORI, an AI tutor on mobile phones, with whom Eedi is co-developing an offline-first DQR app. Comparable as a mission-driven AI tutoring deployment in cost-sensitive emerging markets.

Market position

Strengths5 records

Weaknesses5 records

Competitive moat5 records

Key risks7 records

Key highlights7 records

Customer concentration

Eedi social profiles

Digital presence

Eedi compliance and trust

Trust signal

Compliance1 record

Eedi financial estimates

Financial estimate

Revenue estimate

Valuation estimate

Eedi leadership team

Management profile

Number of profiles

Profiles4 records

Eedi funding detail

Funding detail

Funding overview

Funding rounds3 records

Investors5 records

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

Eedi M&A and investment

M&A and investment

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about Eedi

What does Eedi do?

Eedi builds and operates an AI-powered diagnostic engine for mathematics that identifies and resolves individual student misconceptions in real time using a library of 60,000+ mapped questions and proprietary Knowledge Tracing models. The company delivers this through three product lines: Eedi School (free classroom and homework platform for teachers, students, and parents), Eedi Inside (OEM/embedded licensing of the diagnostic engine for publishers, frontier model providers, and learning platforms), and the Constrained AI Tutor (human-in-the-loop tutoring that activates when diagnostic checks are failed).

Is Eedi a public or private company?

Eedi is a private company. It is classified as venture growth investor backed and is currently operating.

When was Eedi founded?

Eedi was founded in 2015. It employs 51 to 100 people.

Where is Eedi based?

Eedi is headquartered in London, United Kingdom, in the Europe region.

How does Eedi make money?

Four revenue lines are on record. School/Teacher Subscriptions (Free Tier) is the primary driver. The others are eedi School Plus (Parent/Student Subscription), intelligence Infrastructure / Licensing and partnership / Integration Revenue.

Who are Eedi's main competitors?

Direct peers on record are Imagine Learning, Khan Academy, Zearn, IXL Learning, DreamBox Learning, Carnegie Learning, Mathspace and Squirrel AI. Quizlet is listed as a broad incumbent. Rising Academies is listed as an emerging player.

Does Eedi have an API?

No public API is recorded for Eedi.

What industry is Eedi in?

Eedi's product category is EdTech AI Infrastructure. Its primary akta.pro industry code is EDAFACAF, Digital Labs, Simulations & Virtual Practicals (Non-XR). Its NAICS code is 611 and its SIC code is 8200.

Unlock the full company data

50 free credits on sign-up, no credit card required.

Contact sales
Live signals
YahooFive Things You Should Know About New AI Tutor StudyEedi and Google DeepMind ran a trial of 1,179 students in 10 U.K. schools, finding AI tutors matched expert human tutors in fixing mistakes. However, neither improved standardized math scores, and tutored boys performed worse than static-practice boys. Eedi is releasing subgroup data and results on Thursday.FutureEdResearch Notes: Two Emerging Strategies for Using AI in TutoringTwo new randomized controlled trials found that AI embedded in chat-based math tutoring can improve student academic outcomes, raising questions about tradeoffs between cost and the value of personal connections provided by human tutors. A study by Google and Eedi Labs evaluated LearnLM, where supervising tutors approved 76.4% of AI responses with little or no edits and students using LearnLM outperformed those tutored by humans (66% vs 61% success rate). A separate Stanford University study found that Tutor CoPilot, an AI tool assisting human tutors, increased student topic mastery by 4 percentage points, with gains of up to 9 points for lower-rated tutors, illustrating two contrasting approaches to integrating AI in education—one substituting human tutors and one augmenting them.Stanford University School of EngineeringResearch Notes: Two Emerging Strategies for Using AI in TutoringTwo randomized controlled trials published findings on the effectiveness of AI in K-12 math tutoring, with one study conducted by Google and Eedi Labs evaluating their LearnLM system among 165 students, and another by Stanford University testing Tutor CoPilot among 1,000 students. The LearnLM study found AI tutoring matched or exceeded human tutors (66% vs 61% success rate) and outperformed static hints (56%), while the Tutor CoPilot study showed a 4 percentage point improvement in topic mastery, with gains up to 9 points among lower-rated tutors. Both studies suggest AI could help scale tutoring at lower cost while maintaining or improving learning outcomes.YahooAI tutors, with a little human help, offer ‘reliable’ instruction, study findsEd-tech startup Eedi.com conducted a randomized controlled trial with 165 British secondary school students (ages 13-15) in which an AI-powered tutor using Google's LearnLM LLM, paired with human oversight, slightly outperformed human-only tutoring, with students solving new problems successfully 66.2% of the time versus 60.7% with human tutors alone. The supervised AI tutor hallucinated just 0.1% of the time across 3,617 messages and enabled students to correct misconceptions over 90% of the time, suggesting that pedagogically fine-tuned AI given detailed student proficiency data could deliver effective personalized tutoring at scale. The findings indicate AI could serve as a front-line tutor with humans intervening only when needed, potentially addressing accessibility limitations that require hiring large numbers of human tutors.PR NewswireImagine Learning and Eedi Launch the First Curriculum-Embedded AI Assessments for U.S. Math ClassroomsImagine Learning announced a strategic partnership with Eedi Labs to integrate real-time, AI-powered diagnostic assessments directly into Imagine IM, their core math curriculum, launching for back-to-school 2026 for U.S. classrooms. The Eedi diagnostic engine uses over a decade of student learning data and machine learning to identify student misconceptions in as few as 10 questions, allowing teachers to adjust instruction immediately without additional testing tools. The partnership aims to extend the real-time assessment model beyond math to other subjects using Imagine Learning's portfolio of ELA and science solutions.LelezardImagine Learning and Eedi Launch the First Curriculum-Embedded AI Assessments for U.S. Math ClassroomsImagine Learning announced a strategic partnership with UK-based Eedi Labs to integrate AI-powered, real-time assessments directly into its Imagine IM math curriculum, launching for the 2026 school year. The system uses Eedi's diagnostic engine, built on over a decade of student learning data and machine learning, to give teachers immediate insight into student misconceptions and enable faster instructional adjustments. The partnership aims to expand real-time assessment capabilities across Imagine Learning's broader portfolio of K-12 education solutions.EdsurgeTeaching Machines to Spot Human Errors in Math AssignmentsEedi Labs, a U.K.-based edtech platform, partnered with The Learning Agency and Vanderbilt University to run a Kaggle competition that used AI to predict student math misconceptions, winning $55,000 in awards from the Gates Foundation and Walton Family Foundation. The competition achieved "impressive" accuracy in identifying misconceptions from multiple choice questions and student explanations, building on Eedi Labs' existing AI tutoring platform. A randomized controlled trial conducted with Google DeepMind found that Eedi's "human in the loop" AI tutoring boosted learning in 11-12 year olds in the U.K., and the platform is now used by 4,955 students across 39 K-12 schools and tutoring networks in the U.S., with another trial planned with Imagine Learning in 2026.MorningstarNew Exploratory Research From Eedi and Google DeepMind Reveals Human-in-the-Loop AI Tutoring Outperforms Human-Only SupportEedi and Google DeepMind tested a human-in-the-loop AI tutoring model in five UK classrooms, finding it matched expert human tutors in immediate correction (93.0% vs 91.2%) and misconception resolution (95.4% vs 94.9%). The AI doubled the human tutor's impact on knowledge transfer, boosting learning by 10 percentage points versus 4.5. A 2026 U.S. RCT with Imagine Learning is planned.Business Wire BlogNew Exploratory Research From Eedi and Google DeepMind Reveals Human-in-the-Loop AI Tutoring Outperforms Human-Only SupportEedi and Google DeepMind tested a human-in-the-loop AI tutoring model in UK classrooms, finding it matched human tutors in immediate correction (93.0% vs 91.2%) and misconception resolution (95.4% vs 94.9%). The AI doubled knowledge-transfer gains, improving learning by 10 percentage points versus 4.5 for human tutors alone. A 2026 U.S. RCT with Imagine Learning is planned.EediNew Exploratory Research from Eedi and Google DeepMind Reveals Human-in-the-Loop AI Tutoring Outperforms Human-Only SupportA new exploratory study by Eedi and Google DeepMind, conducted in summer 2025 across five UK secondary school classrooms with 165 students and 17 tutors, found that a human-in-the-loop AI tutoring model matched or exceeded the effectiveness of expert human-only tutoring, with the AI-human team achieving a 10 percentage point improvement in knowledge transfer compared to 4.5 points for human tutors alone. The trial used LearnLM models from Google and Google DeepMind on Eedi's platform, with AI-generated messages showing zero harmful content and 82.3% approved by expert tutors with minimal edits. The success has led to plans for a 2026 randomised controlled trial across U.S. districts in partnership with Imagine Learning, testing the model's scalability and long-term learning gains.