Intelligence that reasons.
AINA neither improvises nor guesses. It analyses sources, cross-checks facts and lays out the traceable reasoning behind every conclusion. Whenever a piece of data is unverified, it states so openly.
AGI · AINTECH Global Intelligence
From
intention
to
result.
Artificial intelligence agents engineered to execute real work in your company: every answer backed by verifiable proof, every action subject to your explicit permission, and your data exactly where you decide. Today, in early access.
4
AINA One editions
EU
Your data, in Europe or on your own infrastructure
€0
To export or delete your data
0
Cookies and trackers
Scroll
Manifesto
Four promises. One way of working.
We do not sell opaque promises. We build intelligence systems that can be explained, governed and audited end to end, respecting at all times who is in command: you.
Technology that connects.
AINET orchestrates your existing tools, your data sources and your teams within a single intelligent operational network. Without forcing you to replace what already works: making it work together with precision.
Security that protects.
Every operation with real-world impact requires your prior confirmation. Every piece of data retains its ownership and classification. Every decision leaves an immutable record. That is how an organisation is rigorously protected.
Sovereignty that empowers.
Your information resides exactly where you determine: in European data centres, on your own private infrastructure or in fully air-gapped environments. European sovereignty by architecture.
Platform
Four layers. One system.
Our symbol is the map of the platform. Tap a layer — or hover over it — and watch it light up in the symbol in the background.
The metastructure
The constitutional framework that governs the entire platform: security policies, data classification, regulatory compliance and quality gates. It is the non-negotiable plane of the system that guarantees the reliability of everything else.
The network
Securely integrates your applications, document repositories and teams. Routes each task to the optimal model, prioritising local and European execution without transmitting information externally without explicit authorisation.
The agent
The executive intelligence you and your organisation interact with. It reasons over real context, plans missions, executes tasks and attaches the evidence backing every delivery. Executive rigour available at any hour.
AINCORE The engine
The proprietary core engine on which the entire ecosystem runs: it selects the right intelligence per task, executes it under your jurisdiction, preserves the lineage of every piece of data, and blocks any closure lacking proof.
Three pillars
Technological metastructuregovernance · security
Cognitive capabilitiesreason · remember · choose
Agentic intelligenceplan · execute · prove
How it deploys
CloudPrivateHybridSovereignOn-deviceAir-gapped
From managed service to complete isolation. The same platform, six levels of control.
AINCORE · the engine
The engine that makes everything else possible.
AINCORE is not a thin wrapper around someone else’s model: it is AINTECH’s proprietary engineering. It determines which intelligence resolves each operation, executes it in the environment you mandate, preserves the source lineage of every piece of data, and prevents any delivery from closing when verifiable proof is missing. AINA, the AINET network and every product edition work because this engine sits underneath.
Sovereign routing
Every task goes to the right model, not the most expensive one. First the one that runs on your own hardware, then the European one, and the most powerful models only when they are genuinely needed; if they are outside the EU, only with your prior authorisation. Cost stops being a fixed toll and becomes a decision, task by task.
Local first, then EuropeanMemory with provenance
Every piece of data keeps where it came from, who verified it and when it stops being true. What expires flags itself and drags along everything that relied on it. An organisation that neither forgets by accident nor remembers too much.
Origin and expiry on every piece of dataChain of evidence
Every claim travels with the way to check it again. Decisions are recorded in a chain: rewriting the past breaks the chain and shows. Auditing stops being a favour and becomes a query.
Exportable record for auditsGates that block
No task is accepted as complete without effective verification: if the required evidence or human confirmation is missing, the gate deterministically blocks closure. Not for the machine, not for us.
A check before every deliveryIt is the difference between an intelligence that answers fast and one that can be audited.
AINA · The agent
An agent that shows its evidence.
AINA is not a generic conversational assistant. It is a corporate agent built to strengthen the company across its most demanding areas — strategy, technology, security, finance and product — under one unbreakable rule: before claiming, it verifies; before acting, it requests authorisation. Strategic and operational decisions remain exclusively yours.
StrategyTechnologySecurityFinanceProductSecond opinion
Try it: pick a question and see how AINA checks what it claims and asks for your confirmation when it matters.
What AINA is designed to do
What matters when your business is on the line
AINA
Shows the certainty level and source of every answer
By design
Asks for your confirmation before sending, publishing, deleting or paying
By design
Your data can stay on your own infrastructure or in an air-gapped deployment
By design · six modes
Picks the model per task, local and European first
By design
Memory with an owner, visibility and expiry
By design
Exportable decision log for audit
By design
Says "I don't know" instead of making it up
By design
Design goals of a product in early access, not production measurements.
Worked example, in hours
Repeatable task
Before
With AINA
Preparing a batch of 300 renewals
≈ 20 h
≈ 4 h + 1 confirmation
Tracking 40 issues with deadlines
≈ 12 h
≈ 3 h
Annual compliance calendar
≈ 36 h scattered
≈ 4 h
Illustrative examples with stated assumptions, not client measurements. The assumptions (batch size, people involved and starting hours) are provided in writing to anyone who asks.
Products
One intelligence. Four editions and one platform.
AINA One grows with you — from chat to the whole company — without changing platform. AINET Platform is the same network, opened up for you to build on.
AINA One · Chat
The way in. Ask, delegate, decide — and always see what each answer rests on. In the cloud, from your browser with nothing to install; also in private, hybrid or sovereign deployment.
CloudPrivateHybridSovereign
Certainty labels
Verified, pending or unavailable. Never an answer without its confidence level.
Human confirmation
Send, publish, delete or pay: always with your prior approval.
Extension and bot
Inside your browser and your messaging channels.
Model choice
Local and European first; the most powerful only when needed.
Early accessRequest a seat
AINA One · Personal
Your agent, with memory. It organises, drafts, searches and decides with you, across all your surfaces and, if you need it, inside your own infrastructure.
CloudPrivateHybridSovereign
Its own memory
Remembers what matters and forgets what you ask it to. You see what it keeps.
All your surfaces
Desktop, mobile, browser, Office and voice. The same AINA everywhere.
Proof in every answer
Every claim carries its source. No guessing.
Privacy by default
Private or sovereign deployment for what must never leave.
Waiting listJoin the list
AINA One · Work
The agentic workspace. Spaces, missions, Studio for documents and a shared inbox: your team and AINA working on the same thing.
CloudPrivateHybridSovereign
Missions
Assignments with a goal, a deadline and a success criterion. AINA executes them and reports with proof.
Spaces
One place per project, client or department, with its own permissions.
Studio
Documents, spreadsheets, presentations and code, generated and versioned.
Evidence and decisions
A record of what was decided, when, why and on what basis.
Waiting listTalk to us
AINA One · Business
The whole company. The organisation’s memory and continuity, with administration, roles and compliance: so nothing depends on who happens to be there that day.
CloudPrivateHybridSovereign
Living library
Documents, contracts and knowledge with their version and validity.
Memory with provenance
What is remembered, who can see it and when it expires. All visible.
Administration and roles
Onboarding, permissions, audit and retention, defined by you.
Continuity
Handovers, departures and growth without losing the thread.
In design · waiting listJoin the list
AINET Platform
For those who build. The network: API, SDK, connectors and sovereign routing to embed AINA in your own products, with the same guarantees of control and evidence.
CloudPrivateHybridSovereign
Experience API
Conversation, missions and evidence as a service.
SDK
Integrate AINA into your own application, with the same guarantees.
Open connectors
Designed to be compatible with the standard tool protocol for agents.
Sovereign routing
The same routing: local and European first; the most powerful models only when needed.
Waiting list · developersSee the planned API
SpacesRenewals 2026IssuesNew clientsMissionsAutumn campaignConsent auditLibraryContractsLegal templates
Prepare the renewals mission and tell me what's missing.
Mission createdGoal, deadline and success criterion defined. One thing missing: the send date. Pending
Q4 Renewals · Portfolio72 %1,240 clients · 3 steps · evidence attachedVerified
Once you confirm the date, I'll schedule the send. I won't do it without your approval. Your confirmation
AINA One · Personalon-device
Good morning, Carmen. Three things today.
PriorityThe Talleres Ruiz contract expires on Friday. Renewal drafted, ready for your review.
VerifiedBoth open issues have a reply from the supplier. No action needed from you.
MemoryI'll remember this client prefers to be contacted on WhatsApp. Correct?
Ask or delegate…
Illustrative mock-ups of the design in progress: the names, clients and figures shown are fictitious. The product is in early access and is not yet on sale.
Planned surfacesDesktopMobileBrowserOfficeCode editorTerminalMessaging botVoiceAuthenticator
Verticals
When an industry needs a platform of its own.
The four editions of AINA One serve any company. A vertical is AINA One Business with the trade built in: the same engine, dressed for a specific sector and modelled on the real domains of your organisation.
By vertical
One module per industry, with its language, its processes and its regulation built in. Not a template to fill in: the platform with the trade already inside.
AINA One Business · InsuranceFirst vertical · born inside the sector itself
Professional servicesIn design
Industry and distributionIn design
Public sectorIn design
Insurance is the first vertical and the only one in development today, because we had the problem at home; it will be offered with AINA One Business. The rest are declared design lines, not available product: when they are, this very page will say so.
Pricing
How we charge, without small print.
Prices will be published with general availability. Until then, here is what you can know today: how the Chat, Personal and Work editions and AINET Platform are expected to be charged, and what each includes. AINA One Business and the verticals are still in design; their charging model will be published here when they leave design. It is a planned model: it is not an offer and binds no one until there is a contract.
Individuals · Chat and Personal editions
Early access listPer person, per month
A flat fee per user. No artificial conversation limits and no end-of-month surprises.
- Own memory and all surfaces
- Proof in every answer
- Cancel anytime
Teams · AINA One Work
Talk to usPer team, usage included
A fee per workspace with a volume of missions included; overage is flagged before it is charged, never after.
- Spaces, missions and Studio
- Evidence and decision log
- Optional private or sovereign deployment
Platform · AINET Platform
Developer accessPer use, with an agreed cap
API and SDK billed on real usage, with a monthly cap you set. When it's reached, AINA stops and tells you; it does not keep spending.
- Connectors and vertical modules
- Free test environment
- Partner programme
What we will never charge for: exporting your data, deleting it, or leaving.
Solutions
Built for sectors where mistakes are expensive.
Insurance mediation
Pilot programmeThe first module was born at home.
Case files, renewals, issues, mandatory training and regulator deadlines. AINA knows it because it lives it every day.
- Portfolio and expiry tracking
- Issues with full traceability
- Communications with verified consent
Law firms and advisors
Deadlines, case files and confidentiality.
An agent that knows what it can and cannot say, and keeps a record of every step.
- Case files with evidence
- Client data on your infrastructure
- Drafting with mandatory human review
SMEs and family businesses
Management support without adding headcount.
Strategy, finance, technology and product in one conversation, with the rigour of someone whose company is on the line.
- Decisions with their basis and their risk
- Continuity when the team changes
- Growth without switching tools
Regulated sectors
Control and audit, not promises.
Banking, healthcare, energy, government: where every action needs an owner and a record.
- Sovereign or air-gapped deployment
- Exportable decision trail
- Human confirmation for critical actions
Developers and integrators
DevelopersBuild on AINA.
API, SDK and standard connectors with the same guarantees: every call with evidence, every action under control.
- Planned integration with API, SDK and standard connectors
- Reusable vertical modules
- Partner programme
Your sector?
Describe your caseIf mistakes cost you, let's talk.
Tell us your case. We'll tell you honestly whether AINA works for you today, tomorrow or never.
How it began
Nobody set out to found an artificial intelligence company.
AINTECH came out of a work problem, and each piece appeared because the previous one fell short.
- 01
A real business, not a lab
AINTECH was born inside a working insurance intermediary: clients who call, policies that expire and rules that allow no creative interpretation. When a tool fails there, a metric does not drop: a client gets upset.
- 02
The AI worked; the cost did not
Artificial intelligence solved real tasks — drafting, checking, preparing — but always asking the biggest model did not pay off. Hence the decision that still governs the platform: choose the right model for each task, local and European first.
- 03
Memory and proof
What was missing is what a demo does not show: a system that remembers with provenance and answers that can be checked. No proof, no delivery.
- 04
Today
Four editions of AINA One, a platform to build on and a first module for the insurance sector. Nothing is in commercial operation yet: early access is open to applications for those who prefer to join while things can still change.
Tools
Two answers before talking to anyone.
Which deployment mode suits you and how many hours you could recover. With assumptions in plain sight, like everything else.
Deployment configurator
Three questions and this site proposes the mode and the safeguards it activates, using fixed rules and no AI. It is a guide: it does not replace your own risk analysis.
Recommendation
Choose the options to see the proposal.
Hours calculator
How many hours of repeatable work you could recover per year.
Hours recovered per year
—
Equivalent to — working days.
Assumptions: 46 working weeks · 8-hour day · the share taken on is an estimate, not a measurement.
The Paradigm
More than a platform. A paradigm shift.
We are not here merely to launch another software product or gather a passive community: we are driving a technological movement open to individuals, independent professionals, creators, developers and enterprises alike. Against opaque artificial intelligence that captures your data and demands blind faith, we open the technical vanguard: intelligence that belongs to the end user, proves every claim, and always respects your will.
The old model · Extractive black-box AI
Asking the machine for permission
- Answers without verifiable sources: you bear the risk whenever the AI hallucinates.
- Your conversations, documents and creations feed someone else’s models by default.
- Total dependency on an opaque cloud outside your jurisdiction.
- Built as a closed oracle where the human user is the product.
The New Paradigm · AINTECH Global Intelligence
Intelligence under your command
- Verifiable evidence: every claim travels with its source and its level of certainty.
- End-user sovereignty: your data and memory are yours; they never train any model without your paid written authorisation.
- Confirmed execution: nothing irreversible happens without your explicit approval.
- Open to everyone: from an individual on a phone to a critical air-gapped organisation.
Interact now · Built for every kind of end user
Who are you today? Pick your profile and experience how the paradigm works in your daily life.
AINA One · Chat & Personal · Everyday life
Your personal intelligence, with its own memory and absolute privacy.
Review a lease or mortgage before signing, organise your medical or tax paperwork, plan your week or compare complex purchases without advertisers or external models tracking your private life.
Contracts & paperworkPersonal financeHealth & wellbeingDaily organisation
Live simulation · CitizenTotal privacy
“I uploaded the insurance quote and the bank contract. Which fine print works against me?”
AINA · Document cross-check: I found 2 tie-in clauses on pages 4 and 9. Here is the exact paragraph, the real 5-year cost and a draft reply to renegotiate them. Your files never left your private space.
AINA One · Personal & Studio · Independent professionals
Multiply your output without surrendering the intellectual property of your work.
Lawyers, consultants, architects, physicians, designers and creators: draft reports with verified citations, prepare client proposals, analyse regulations and version deliverables in Studio knowing your expertise will never train a competitor’s model.
Sourced reportsClient proposals100% your IPPer-client memory
Live simulation · ProfessionalProtected work
“Prepare the client opinion cross-referencing their three prior files and current regulation.”
AINA · Studio: Opinion structured in 4 sections. 14 claims linked to their source document and 1 item flagged as Pending client confirmation before issuing.
AINA One · Chat & Studio · Science, study & research
Learn and research on verifiable facts, never on hallucinations.
Whether you are preparing for competitive exams, writing a thesis or researching a scientific field, a fabricated citation ruins weeks of work. AINA separates proven facts from uncertainty at a glance and shows the exact trail of every source.
Paper synthesisVerified citationsTheses & examsStep-by-step reasoning
Live simulation · ResearchScientific rigour
“Compare these five papers and tell me which conclusions have solid experimental backing.”
AINA · Chain of evidence: 3 conclusions have direct experimental replication; 2 are preliminary single-author hypotheses. Comparative table attached with exact page and citation.
AINA One · Work · Teams & SMEs
Your team and AINA coordinated on the same missions and documents.
Leave behind scattered threads and disconnected spreadsheets. Create spaces per project or client, delegate missions with a deadline and success criterion, and keep a shared inbox where humans approve every key dispatch.
Project spacesDeadline missionsShared inboxHuman confirmation
Live simulation · TeamUnder your control
“Coordinate the monthly close with finance and prepare the client notices.”
AINA · Team mission: Monthly close 100% reconciled. All 48 notices are ready in the shared inbox awaiting the manager’s confirmation click to be sent.
AINET Platform · Developers, builders & architects
Build agentic applications with sovereign routing and a hard spending cap.
Embed the AINCORE engine via API, SDK and open connectors. Route to local and European models first, enforce evidence in responses by JSON contract, and set a monthly euro cap that is never exceeded.
Missions APILocal/EU routingStandard connectorsHard spend cap
Live simulation · Builder€0 Sandbox
POST /v1/missions · data_policy: "eu_only" · confirm_before: ["send","pay"]
AINCORE · Router: Task resolved on European node with optimal latency, evidence attached in payload, and automatic gate lock prior to firing external webhook.
AINA One · Business & Verticals · Enterprises & critical sectors
Institutional continuity, lineage memory and deployment all the way to air-gapped.
For insurance intermediaries, law firms, banking, healthcare, industry and public administration: role-based governance, exportable decision audit trails and full operation on your own private or air-gapped infrastructure.
Sovereign deploymentAir-gapped modeAuditable trailIndustry verticals
Live simulation · SovereigntyFull isolation
“Quarterly audit of decisions and portfolio data expiry.”
AINCORE · Chain of evidence: 100% of operations cryptographically chained. Provenance and retention report exported in open format for the compliance committee.
Technical vanguard · Open R&D
The four engineering disciplines that power the movement.
01
Computational epistemology
We replace blind probability with explicit truth states: every output distinguishes what is verified in primary sources from pending inference.
Graded certainty by design
02
Neuro-sovereign orchestration
Dynamic per-task routing that prioritises on-device and European compute before invoking massive external models.
Minimum cost · Maximum sovereignty
03
Lineage & expiry memory
A memory graph where every piece of data retains its author, read permissions and automatic expiry date with cascading propagation.
Verifiable forgetting
04
Deterministic action brakes
Engine-level safety gates that intercept any external effect — send, delete, sign or pay — until explicit human confirmation is received.
Zero unauthorised effects
We build AINTECH by listening directly to the people who will use it every day. Tap the capabilities that matter most to you to register your priority and open a direct line with us:
- AINA on my phone & personal computer (private daily use with zero tracking)Individuals, families & freelancers · Own memory and document review
- 100% local execution on my own hardware (no cloud dependency)Privacy enthusiasts, researchers & professionals under confidentiality
- Studio for documents, spreadsheets & code with verified sourcesCreators, consultants, students, engineers & firms
- Open API, SDK & connectors to build my own sovereign agentsDevelopers, startups, integrators & technical community
Tap “Boost” on any line to add your vote in this browser (zero cookies, zero trackers).
01 · Pioneer Users (Early Adopters)
Test AINA before general availability
We are looking for end users of every profile — not only companies — who want to put AINA to the test in their real daily tasks and shape how every surface evolves with us.
02 · Co-design & direct feedback
Talk directly to the people building it
No middlemen and no support bots brushing you off. Every feature request, use case or technical critique goes straight to engineering and leadership.
03 · Cognitive Sovereignty Manifesto
A European movement with global reach
We hold that artificial intelligence must expand human freedom and judgement, never replace them. If you share that standard, this is your home.
Trust
Security, control, transparency and care.
These are not values to hang on a wall. They are mechanisms designed to act in every conversation and every action.
Security
Data classified by sensitivity, secrets kept out of the AI’s reach and destructive actions blocked without your confirmation.
Control
Every action with an effect on the world — send, publish, delete, pay — waits for your confirmation. And before anything irreversible, it warns you.
Transparency
Every answer shows its certainty level and its source. Verified, pending and unavailable are told apart at a glance.
Care
The minimum that suffices, done well. No unnecessary dependencies, no duplication, no complexity nobody asked for.
GDPR · LOPDGDD
Data protection by design. Personal data classified and not disclosed to third parties except where legally required.
EU AI Act
Transparency about AI use; per-product information sheets, to be published with each launch.
Other frameworks
ENS, NIS2 and DORA: design references, with no certification or conformity assessment.
The frameworks cited guide the design and scope of the platform. AINTECH does not yet hold any certification or formal audit; when it obtains them, they will be published here with their scope, version and date.
Trust centre
Data locationEU by default
Cloud in EU data centres; private, sovereign and air-gapped on your own infrastructure. Using a model provider outside the EU requires your prior authorisation.
Sub-processorsFull list before any test with real data
Each provider with its role, country and guarantees. Any change is notified 30 days in advance.
RetentionYou decide
Periods per data type, verifiable deletion and full export at any time.
Training with your dataOnly if you authorise it, and you get paid
Without your express written authorisation, your data trains no model, neither ours nor anyone else’s. If you decide to provide it for training, you are paid for it.
AuditsNone published yet
They will be published with scope, version and date. We don't show badges we don't have.
IncidentsPublic log when general access opens
What happened, who was affected, how long it lasted and what we changed.
Who we are
We were born where mistakes cost money.
AINTECH Global Intelligence is a Spanish technology company born inside a real business — clients, legal deadlines and zero margin for error — to solve a common problem: artificial intelligence that promises a lot and refuses to be controlled.
Mission
To turn the intentions of people and companies into verifiable results, with an intelligence that reasons, explains and respects its limits.
Vision
To be Europe’s benchmark in sovereign agentic intelligence: systems organisations can understand, control and audit end to end.
Purpose
To ensure that no organisation has to choose between benefiting from AI and keeping control of its business.
Origin
It all started in a real services company. Hundreds of case files, issues, legal deadlines and clients who trust that nobody will make a mistake. That is where an AI that "almost always gets it right" is useless.
We built AINA to work under that demand: show its proof, ask permission before acting, and say "I don't know" when it doesn't. Then we discovered every serious company shares that demand.
Today the AINTECH metastructure, the AINET network and the AINA agent work as one system, under one idea: from intention to result.
AGI stands for AINTECH Global Intelligence: the metastructure that governs, the network that connects and the agent that reasons, working as one.
Value 01
Truth before promise
We only claim what we can prove. Everything else is marked pending, and you can see it.
Value 02
You decide
The AI proposes and executes; decisions with consequences are yours. Always.
Value 03
Total transparency
Every answer shows where it comes from and what supports it. No black boxes.
Value 04
Care in the detail
The minimum that suffices, done well. No unnecessary dependencies or gratuitous complexity.
Value 05
Sovereignty
Your data, in your territory and under your rules. Europe as the standard, not the exception.
Value 06
Honesty
If an idea is weak, we'll tell you first and propose a better one.
Milestone 01
Origin
A real company with a real problem: too much repeatable work, zero margin for error.
Milestone 02
Construction
AINA is born as an internal agent. A doctrine of evidence, human control and sovereignty from day one.
Milestone 03
Platform
The AINTECH metastructure, the AINET network and the AINCORE engine are formalised as one system.
Today
Early access
Early access open to applications and a progressive opening of the four editions.
Team
People who decide. Intelligence that executes.
A small, experienced team: people decide and answer for every delivery; AINA works alongside them. And we are growing.
Leadership
RS
Ramón Soriano Agulló
Founder · Chairman · CEO · CFO
Entrepreneur. In insurance intermediation since 2007 and running an agency since 2020. Built AINA for that work before offering it to other companies, which is why the platform starts with the problems that cost hours, not with the ones that look good in a demo.
EducationDouble degree in Business Administration and Marketing & Sales Management, ESIC. Master’s in Marketing and Sales Management (GESCO), ESIC. Advanced Programme in Financial Advice (CAFP®), Afi. Agency Succession Programme, Complutense University of Madrid.
Public rolesPresident of JOVEMPA Marina Baixa (2021–2023) and vice-president of the JOVEMPA federation. Advisory board of the CdT Benidorm “Domingo Devesa”.
Team
JL
Jorge López Pérez
CTO · CISO
- LauraCCO · Customer care
- AlbertoCXO · Support
- PauCOO · Operations
- ToniCMO · Marketing
- Juan IgnacioCSO · Sales
Advisors
AC
Alfonso Calero Romero
Partner · CLO
JA
José Luis Arbillaga Guerrero
Tax advisor
Open positions
We are looking for people in product, engineering, security and brand.
Roadmap
What's coming, in the order it will come.
We publish the plan because we believe in transparency. The stages and timeframes are indicative and are updated with evidence, not wishes.
Now · 2026
AINA One pilot programme
Internal use and first pilots with selected clients.
Next
AINA One Work for teams
Spaces, missions and Studio for small and medium organisations.
Then
AINET Platform
API, SDK and connectors for partners and integrators.
Horizon
AINA One Business
Organisational memory with provenance, the first vertical (insurance) and a trust centre with published audits.
Developers
How you will start building.
This is how you ask AINA for something from your own code: a mission with a goal, a success criterion and a confirmation policy. The API is in design; this example shows the shape it will have, not a production service.
- Request developer access. We'll let you know as soon as the test environment opens; access and the environment are free of charge.
- Create a mission. Goal, deadline, success criterion and which actions require human confirmation.
- Read the evidence. Every result arrives with its certainty level and source; you decide what to do with what isn't verified.
- Connect your tools. Standard connectors for agents; your data doesn't leave without your policy.
// Illustrative example · API in design
POST /v1/missions
{
"objective": "Prepare Q4 Renewals · Portfolio",
"success": "300 communications ready with verified consent",
"deadline": "2026-11-15",
"confirm_before": ["send", "delete", "pay"],
"data_policy": "eu_only"
}
// Response
{ "status": "awaiting_confirmation",
"evidence": [{ "claim": "1,240 valid consents",
"level": "verified" }],
"pending_action": "send" }
Careers
Open positions.
Small team, fast decisions, high standards. You'll work directly with leadership and with AINA. Remote in Spain, with in-person meetups.
Head of Product Remote · Spain+
Turn the platform into products people understand and pay for. You will define the four editions and the platform, prioritise on evidence and say no often.
We're looking for: experience launching software products to businesses, an obsession with clarity, comfort with uncertainty.
Platform engineering Remote · Spain+
Build and operate the metastructure: model routing, deployments from private to air-gapped, observability and tests that don't get skipped.
We're looking for: distributed systems, security as a habit, and the drive for everything to have a verifier.
Security and compliance Remote · Spain+
Put the trust centre into practice: data classification, controls, formal audits and European compliance (data protection, AI, resilience).
We're looking for: experience in regulated sectors, judgement to tell what protects from what only looks like it does.
Brand and communication Remote · Spain+
Guard the AINTECH brand and tell what we do without exaggerating: web, press, community and evidence-based content.
We look for: precise writing, an eye for visual detail and an allergy to empty promises.
01Write to us
An email with what you've done and why this interests you. No template.
02Conversation
45 minutes with leadership. Real questions, both ways.
03Real test
One of our real problems, paid, with AINA at your side.
04Decision
We aim to decide within two weeks of first contact, and we always explain why.
Questions
What we get asked before starting.
Does my data leave my company?
It depends on the mode you choose. In the cloud, your data is processed in EU data centres; in private, sovereign and air-gapped deployments, it stays in your infrastructure. In every case AINA prioritises local and European models, and sending data to an external model provider or outside the EU requires your prior authorisation.
What does "every answer with its proof" mean?
That every claim carries a certainty label — verified, pending, unavailable — and the source that supports it. If AINA hasn't checked something, it says so instead of making it up.
Can AINA act without my permission?
Not for actions with consequences: sending, publishing, deleting, signing, paying or changing a production system require your explicit confirmation. Before asking, AINA shows you what it will do, who it affects and whether it can be undone; if it cannot, it warns you.
Does it replace my team?
It multiplies it. AINA takes on repeatable work and contributes expert analysis; decisions remain with people. That is why we talk about support that proposes, not a boss who decides.
When can I start?
We're in an early access programme. Request a seat and we'll tell you the real status of each line, without promises we can't keep.
Is it only for one sector?
It was born in a very demanding sector —insurance— and that is our first vertical module, but the platform is general. Early access is open to firms, SMEs and regulated sectors; the developer environment has not opened yet.
Do you use my data to train models?
Only if you authorise it expressly and in writing, and in that case you are paid for it. Without that authorisation, your data trains no model, neither ours nor anyone else’s.
What if AINTECH disappears?
It is the right question and we answer it plainly: we are a young company, in early access. That is why the design does not lock you in: in private, sovereign and air-gapped deployments the system runs on your infrastructure, and in every mode your data and your memory are yours and you can export them in open formats whenever you ask. If we ceased to exist tomorrow, you would keep what you already have on your infrastructure and everything you have exported.
Contact
Tell us your intention. We'll deliver the result.
Tell us what you want to achieve. We'll reply with what we can do today, what's coming, and what we won't do.
Early accesshi@ain-tech.cloudTo try AINA in your company. We usually reply within two working days.
InvestorsDossier under a confidentiality agreementFigures, roadmap and corporate structure.
Work with usSend your applicationTell us what you have built. No templates, no thousand-field forms.
Head officeCalle Constitució, 5 bajo
03570 La Vila Joiosa · Alicante
Spain · European Union
03570 La Vila Joiosa · Alicante
Spain · European Union
HoursMonday to Friday, 9:00–18:00 (CET)
CompanySoriano Costa Blanca Services, S.L.
Tax ID B02779841
Tax ID B02779841
Presshi@ain-tech.cloud
Newsletter
Once a month: what we've built, what we've learned and what's next. No noise.
Early access
Intelligence that can be explained.
If you're looking for an AI that impresses, there are many. If you're looking for one you can control and audit, and trust, let's talk.
Request early access
© 2026 Soriano Costa Blanca Services, S.L. · Tax ID B02779841