For CEOs, COOs and operations leaders who want clear priorities, clear owners and results they can measure.
Adopt AI with rigour, not guesswork.
For heads of localization, producers, brand leads and legal teams in games, publishing and media.
A consultancy built for the move from AI pilot to production.
Our founders spent more than ten years delivering quality under pressure. They saw first-hand how AI changed that work.
KREQI works through two branches, backed by one quality engine. KREQI Corporate helps leaders adopt AI across the business, with clear priorities, clear owners and results they can measure. KREQI Creative helps studios, publishers and brands keep AI content accurate and culturally right in every language. Both branches rely on the same native linguists, test sets and scoring method, and hold every project to the same standard of proof.
Our founders come from senior quality assurance and localization roles in the games industry, one of the most demanding software sectors. This shapes how we work. We treat AI output like any software. We measure it, validate it, and trust it only once it has proven reliable.
We keep the firm lean on purpose. Senior consultants lead every engagement. We bring in vetted specialists and professional linguists for each project, so you never pay for an idle team. You get the experience you hired and the flexibility your project needs.
We work remotely and on site across Europe, from Warsaw and Lisbon.
Practitioners, not observers
We have managed quality and delivery under real deadlines. Our advice works in production.
Quality in everything we do
We measure every engagement. If we cannot show the difference, we do not claim it.
People first
AI adoption changes how people work. We design for the teams who use it every day.
Adopt AI with rigour, not guesswork.
What we do, and how we do it.
Help companies succeed with AI, responsibly.
We turn AI plans into working, measurable capability. We stay honest about what AI can and cannot do. We build the quality controls that make it safe to rely on. And we leave every client more capable than before.
A market where AI adoption gets measured, not just marketed.
We want organisations of every size to adopt AI with the same care once reserved for critical software. Evidence comes before hype, and people stay in control.
Rigour over hype
We recommend what the evidence supports, even when it sounds less exciting.
Clarity
Plain language, a defined scope and results you can hold us to.
Partnership
We build skills within your team, not dependence on ours.
Responsibility
We weigh risk, ethics and people alongside performance.
Craft
We treat quality as a discipline, not a checkbox.
Measurable value
Our work is done only when we can show the difference we made.
Make AI work across your business.
For CEOs, COOs and operations leaders who want clear priorities, clear owners and results they can measure.
AI Readiness Check
We review your goals, data, skills, risks and ownership. You get a readiness score and your three best starting points. About 2 weeks.
Pilot Rescue
A keep, fix or stop verdict on AI pilots that go nowhere, with the cost and the value of each use case.
AI Strategy Roadmap
We score every AI idea on value, risk, quality, adoption and law. We keep the best three to five, each with a target and an owner.
AI Responsibility Map and Policy
First, we list every AI tool and name one owner for each. Then we write your AI policy with your teams: tools, data, uses and sign-off. You get a proof file for your board or insurer, a short policy and a one-page staff version.
AI Training for Staff
Short sessions by role on how to use, check and correct AI, with the training record the EU AI Act expects.
Advisory and Leadership
A senior advisor at your side. A base retainer plus an outcome-linked part, agreed up front.
AI Sounding Board
A quarterly review of your AI targets and governance, plus questions by email between meetings.
Also available: a part-time Head of AI Quality, two days a month, for mid-sized companies.
Native linguists, test sets and a clear scoring method. Each branch uses it with its own tests.
Multilingual AI Quality Scorecard
We test your customer support bot in several languages. Native linguists grade every answer, and each language gets a score.
AI Output Monitoring
Monthly tests of your AI answers per language, and a full check before every model change.
EU AI Act Check
Do your chatbots tell users they are talking to an AI, as Article 50 requires? A scored check and a fix list.
AI Review Workflow Design
Risk-based review steps that stop poor AI work from spreading through your teams.
Start with an AI Readiness Check.
A short, fixed-scope first step that shows where you stand.
Keep AI content right in every market.
For heads of localization, producers, brand leads and legal teams in games, publishing and media.
Human Translation Check
We check whether a translation sold as human really was, through invoices, work files and the text itself. Each language gets an A to E grade.
Cultural Assurance Report
Native reviewers check the tone, cultural fit and brand voice of your content in each market. We also check how public AI tools describe your brand.
Games AI Rights Review
Voices, art and text. We check actor consent, ownership of AI-made assets and the AI disclosure on your store pages.
Independent Vendor Audit
On the buyer's side, we check how your translation and testing vendors really work, including hidden AI use. Also for investors in due diligence.
Games AI Governance
Provenance logging, disclosure workflows and content marking for AI assets, in line with the EU AI Act and platform rules such as Steam.
Independent Quality Authority
We act as the independent quality layer above your LQA and QA vendors. We set the standards, audit the work and check cultural risk at the design stage.
Localization AI Governance
We set the rules for AI in your localization pipeline. Quality thresholds per language and content type, clear escalation paths, and the points where a human must review. Then we help your teams and vendors apply them.
The A to E grade
One letter per language, easy to read. It can cover each part of a product, for example translation A, art B, voices A.
Native linguists, test sets and a clear scoring method. Each branch uses it with its own tests.
Multilingual AI Quality Scorecard
We test in-game AI characters, NPC dialogue and AI-written text in every language you ship.
AI Output Monitoring
Monthly tests of your AI content and characters, with alerts when quality drops in a language.
EU AI Act Check
AI voices, AI images, deepfakes and content marking. A scored Article 50 check and a fix list.
AI Review Workflow Design
Review steps for localization and content teams, so reviewers check the right things without slowing down.
Start with a Human Translation Check or an AI Act Check.
A short, fixed-scope first step that shows where you stand.
Clear engagement models.
Each branch follows the same ladder: a short first step, a main project, then a monthly service. You know the price before we start.
KREQI Corporate
AI Readiness Check, Pilot Rescue
Strategy Roadmap, Responsibility Map and Policy, Training, Scorecard
Advisory and Leadership, Sounding Board, Output Monitoring
KREQI Creative
Human Translation Check, EU AI Act Check
Cultural Assurance, Rights Review, Vendor Audit, Games AI Governance, Localization AI Governance
Independent Quality Authority, Output Monitoring
Prices are indicative and confirmed in a written proposal after scoping. Larger programmes can combine several models in one agreement.
Media and resources.
Everything journalists, partners and event organisers need to write about KREQI accurately and reach the right person quickly.
KREQI launches to help European companies validate AI quality
CompanyIntroducing the AI Readiness Assessment, a fixed-scope first step into AI adoption
ProductWhy LLM quality validation has become a board-level concern
PerspectiveAbout KREQI in brief
KREQI is an AI quality and cultural assurance consultancy based in Warsaw and Lisbon. It works through two branches. KREQI Corporate helps companies adopt AI with clear priorities and owners. KREQI Creative helps games, publishing and media teams keep AI content right in every language.
Media contact and brand assets
Press enquiries
press@kreqi.com
Brand assets
Logo, wordmark and colour references are available on request. Please keep clear space around the logo and do not change its colours.
Spokespeople
Our founders can comment on AI adoption, LLM quality and localization.
Work with us.
We work with senior consultants, linguists and specialists across Europe. We bring people together for each project, so you work on topics that match your expertise.
Linguistic Quality Reviewer
Review AI and human content in your native language. Grade chatbot answers, flag errors and explain what to fix.
What you bring
- Native speaker of a European or Asian language
- 3+ years in translation, localization or LQA
- Precise, consistent written feedback
LLM Evaluation Specialist
Design test sets, run evaluations and turn results into clear quality scores for our clients.
What you bring
- Hands-on experience evaluating LLM output
- Comfort with data, scoring and test design
- Fluent English, a second language is a plus
Cultural Specialist, Games and Media
Review games and media content for cultural fit and market risk in your region, from symbols to history and humour.
What you bring
- Deep knowledge of one market's culture and history
- Experience in games, media or publishing
- Clear, well-argued recommendations
Spontaneous application
No role that fits? Tell us about your expertise. We keep every profile and contact you when a matching project starts.
careers@kreqi.com
Let's talk about your AI plans.
Tell us the outcome you are looking for. We will tell you clearly whether and how we can help.
Lisbon, Portugal
This opens a pre-filled email in your mail app.