Trusted by Industry Leaders
Put XGEN to work — free for 15 days
Every Agentic AI workflow feature, free for 15 days, with none of the setup burden that usually comes with on-premise
Plateer AI Labs
Research becomes product, and the field proves it
We start with foundational research, validate it in open source, and ship it as XGEN, Polar, and AI Code Assistant. Customer deployments and quality certification are where the value gets settled.
- 01
Research
Foundational work on Enterprise AI that is trustworthy, sovereign, and composable
Explore - 02
Open Source
The proven libraries underneath that research, published in the open
Explore - 03
Product
Applied in the enterprise as XGEN, Polar, and Code Assistant
Explore - 04
Proof
Results evidenced by customer cases and national certification
Explore
/ Product Tour
See the products in motion
Short walkthroughs of XGEN and the key capabilities built on it
XGEN
Overview
An enterprise AI agent runtime built on a node canvas and a headless engine — a tour of the XGEN platform end to end
/ Applied AI
Enterprise AI, proven industry by industry
Finance, public sector, commerce, IT services — we research and validate Enterprise AI that understands each industry's workflows and regulations
E-Commerce
AI Commerce Automation
- Automated product review
- VOC analysis and response
- Price monitoring
- Promotion optimization
Public Sector
AI for Digital Government
- Automated citizen support
- Administrative assistance
- Policy and regulation search
- Public data utilization
Finance
Trusted Financial AI
- Credit review support
- Contract review
- Anomalous transaction detection
- Automated regulatory compliance
IT Services
Enterprise AI Operations
- Technical document search
- Incident analysis
- Operations automation
- Coding assistant agents
Enterprise AI you can verify — GS certified
Rather than claiming performance, we had the quality tested by an accredited third party. XGEN holds Grade 1 GS (Good Software) certification — the highest grade available
GS Certification (Good Software)
Grade 1 awarded
A national software quality certification
GS (Good Software) certification is operated by Korea's Ministry of Science and ICT under the Software Promotion Act. Accredited testing laboratories evaluate product quality against the ISO/IEC 25000 series of international standards.
Three areas tested in realistic conditions
Testing covers the product specification, the user documentation, and the running software in an environment close to real use, assessing functional suitability, performance efficiency, reliability, and security.
What the certification unlocks
- Eligible to apply for Excellent Procurement Product designation
- Qualifies for mandatory split-ordering in public software projects
- Designated for priority purchase by the Ministry of SMEs and Startups
- Scored technical credit, backed by accredited third-party testing
/ Research
Research that makes Enterprise AI practical
In the enterprise, AI adoption stopped being a question of model benchmarks a while ago
AI you can trust
Answers grounded in the knowledge and data your company already holds — not in what the model invented
Data sovereignty
AI belongs on top of your internal data. We research on-premise-first architecture to keep it there
Composable, extensible AI
A Composable AI architecture that modularizes agents, workflows, knowledge, and tools so they can be recombined
Open Source
We prove the engineering in the open
The core technology behind XGEN is published as production-grade Python libraries. Use them freely under the MIT license, install withpip, or run them straight from the browser gallery.
- Open-source libraries
- 0
- Categories
- 0
- License
- MIT
- Install
- pip
/ Technology
The engines and frameworks behind real operations
Past the experiment and into production — we research and design the core engines and frameworks Enterprise AI runs on
A knowledge engine that follows the facts
It traces the relationships between your data rather than stopping at vector similarity, and it shapes the whole environment the AI works in
Operational intelligence and retrieval
Precise knowledge context bound to verified execution, so the result holds up
Not a wrapper — it compiles to code
Workflows and policies become standard code, exported as a self-contained package that runs anywhere
/ Live Demo
Try it right now
X2BEE AI and the XGEN Agentic AI Platform both run in your browser — no installation required
/ Insight
Insight from our research and the field
XGeny — from an AI that answers to an AI agent that works
The XGEN agent takes another step forward. XGeny is a new agent runtime with memory and a workspace of its own, able to find the tools it needs — and build them when they do not exist — to carry work forward.
Read the post/ Latest tech notes
Bringing Jaeger to XGEN: from Envoy sidecars to OTel auto instrumentation
Introducing Jaeger tracing alongside an XGEN service migration — running into the limits of the Envoy sidecar, and building distributed tracing with OpenTelemetry auto instrumentation instead.
Read the postWhy we turned the user manual into a solution guide wiki
Generating a web wiki, Word, Korean, and English documents from one Markdown source — and why we now operate the manual like a product.
Read the postSetting up a blog authoring workflow developers will actually use
Letting developers publish from their own GitHub account without opening the main repository — lowering the cost of participation with Open Authoring.
Read the post/ Resources
Technical resources from Plateer AI Labs
Documentation and release notes for Enterprise AI, all in one place
/ frequently asked
Frequently asked questions
- What is Plateer AI Labs? +
- Plateer AI Labs is an enterprise-AI research lab that turns research into products. We open-source the building blocks, build the XGEN Agentic AI Platform on top of them, and validate everything through real customer PoCs — a research → open source → product → proof pipeline.
- What is XGEN? +
- XGEN is an on-premise Enterprise AI platform for designing, deploying, and governing Agentic AI services on the LLMs and infrastructure you choose. Built on a node canvas and a headless engine, it lets teams build agents without code and run them safely with knowledge (RAG), tools (MCP), and governance. It is model-agnostic, so you can mix open-source and commercial models with in-house GPUs by purpose, cost, and accuracy.
- Does it run on-premise / air-gapped, and is our data safe? +
- Yes. XGEN runs on-premise so data never leaves your environment, and it supports network-separated and air-gapped deployments. All AI models and data stay on the internal network, protected by an authentication gateway and trust boundary, RBAC/ABAC access control, PII masking, unified audit logs, and AI risk-grade policies across every layer.
- What does XGEN's GS certification Grade 1 mean? +
- GS (Good Software) certification is Korea's national software quality certification, operated by the Ministry of Science and ICT under the Software Promotion Act. Accredited labs test functional suitability, performance efficiency, reliability, and security against the ISO/IEC 25000 series of international standards. Grade 1 is the highest level — meaning XGEN Agentic AI Platform's quality is verified by independent, accredited testing rather than vendor claims. For buyers, it qualifies XGEN for public-procurement programs (Excellent Procurement Product designation, split-order eligibility, priority purchase) and counts as a scored technical credential in many public RFP evaluations.
- What are the open-source libraries, and how do they relate to XGEN? +
- We open-source the document-ingestion, knowledge-graph, and agent technologies that power XGEN as production-grade Python packages. All are MIT-licensed, free for commercial use, installable via pip, and runnable directly in the browser gallery — research proven in the open, with the product (XGEN) built on top.
- Which industries is it applied to? +
- Finance, public sector, commerce, manufacturing, media, logistics, telecom and more — adapted to each industry's workflows and regulations. See Customer Cases for PoCs and projects validated in the field.
- How do we get started? Is there a free trial? +
- Request a product demo, a 15-day XGEN free trial, or a PoC / technical consultation from a single form. A Forward Deployed Engineer works with you from requirements through design, implementation, and enablement. Pick an inquiry type and we'll reply within 1–2 business days.