15-day free trialStart XGEN right in your browser — nothing to installStart free trial

Solutions · Customer cases

Finance customer cases

Customer cases applying AI to financial work — generative-AI platforms, document review, and search modernization

01Overview

AI for finance work

Financial work comes with strict limits on data export and external model use, and every AI decision and action must hold up to internal control and audit. XGEN is applied by running agents and knowledge search inside the customer's infrastructure, with permissions, approvals and audit logs managed together.

02Challenges

What finance teams run into

  1. 01

    Data and models must stay inside

    Documents and systems holding customer and transaction data are hard to connect to external cloud AI. A configuration that works under network separation has to come first.

  2. 02

    Separate AI projects are hard to govern

    When PoCs scatter across departments, models, prompts and permission rules diverge, and it becomes hard to tell who accessed which data.

  3. 03

    Approval must follow risk level

    Customer-facing answers, internal summaries and system actions carry different risks. Each AI service needs its own review and approval path.

03Approach

How it is applied, step by step

  1. 1

    On-premises platform

    XGEN is installed on the customer's servers and GPUs, connecting in-house LLMs and approved external models under policy.

  2. 2

    Search grounded in internal documents

    Regulations, product manuals and work guides are indexed with RAG, and answers cite their source documents.

  3. 3

    Approval routes and AI risk review

    Each AI service is risk-rated, and deployment is approved through a route of partner teams, governance owners and the information security officer.

  4. 4

    Operation and audit

    RBAC, PII masking, guardrails and execution audit logs track access and actions in operation.

04Deployments

What we have delivered

Company I · Capital finance|In progressXGEN

Building an enterprise AI platform and governance foundation for capital finance

Plateer is building an XGEN-based, on-premise enterprise AI platform that enables Company I to apply AI agents and RAG safely to internal capital-finance work. The engagement also establishes one operating and governance framework across agent development, validation, approval, deployment, operations, and audit.

Challenge

Company I needed a governed foundation for using multiple AI models and internal data while meeting strict financial-sector security and control requirements. It also needed to prevent departmental pilots from becoming disconnected systems and to manage models, knowledge, tools, and agents on one platform. A traceable standard was required across development, validation, approval, deployment, operations, and audit, with consistent access controls inside an on-premise environment.

What we built

Plateer is designing and implementing an on-premise enterprise AI platform based on XGEN and aligned with the customer's infrastructure and security policies. The platform brings together agent design, RAG knowledge pipelines, AI model and tool connectivity, governed validation and deployment, and operational monitoring. RBAC, audit logs, guardrails, and data-protection policies provide consistent control over users, organizations, tasks, and execution records. The current phase establishes a reusable architecture and operating standard that can support additional departments and AI services over time.

Scope delivered

  • XGEN-based on-premise platform design and implementation
  • Agent design, RAG knowledge pipelines, and multi-model connectivity
  • Governed validation, approval, and deployment workflow
  • RBAC, audit logs, guardrails, and data controls
  • Operational monitoring and reusable expansion standards
Read the full case
Jeju Bank|In operationXGEN

Jeju Bank's internal generative-AI platform

Building an internal generative-AI platform on XGEN with Jeju Bank, then stabilizing it through on-site operations and widening its use across the bank

Challenge

Generative AI had to reach internal work while meeting the security and network-separation requirements of the financial sector — and it had to land in real operation, not stop at a one-off build.

What we built

An XGEN-based generative-AI platform was built into the bank's own infrastructure and stabilized through an on-site operating team. On the strength of the first phase, follow-on work is widening where it gets used.

Scope delivered

  • Internal generative-AI platform built
  • Network-separated, on-premise operation
  • On-site operations · follow-on expansion
Read the full case

05Typical work

  • 01Q&A over internal rules and lending manuals
  • 02Drafting responses from consultation history and product documents
  • 03Registering, approving and running departmental AI services on one platform

06Before you adopt

  • Model performance and answer quality depend on the LLM and GPU configuration.
  • Approval routes and risk criteria are defined together with the customer's internal-control rules.
  • Regulatory compliance must be assessed by the customer; adopting the platform alone does not satisfy it.
Finance customer cases · Plateer AI Labs