Abstract
This paper explores the main challenges that government agencies face and suggests a distributed platform architecture for data integration, cleaning, and analysis, incorporating artificial intelligence and machine learning algorithms. It discusses the platform's infrastructure and software core, as well as the idea of using enterprise AI with vector data storage. We describe scenarios where users analyze data using AI agents and where AI agents manage business processes with a controlled feedback loop. The solution is designed for partially or fully isolated environments, using an open-source-first approach and planned capacity allocation.
Introduction
The complexity and fragmentation of government information systems require a method to unify data from multiple sources, enabling analytics and management capabilities that were previously out of reach. The proposed platform creates a unified pipeline for ingesting, storing, cleaning, normalizing, and analyzing data, as well as executing and optimizing processes. It includes enterprise AI that offers contextually relevant answers and secure access to action tools. We focus on practical design choices suitable for single-cluster, resource-constrained, and isolated deployments.
Current Challenges
Based on years of experience building systems for government agencies, we have identified the following priorities:
Data integration and processing. There is a need to combine and process data from fragmented systems to enhance decision-making and gain a complete analytical view that no single system can provide.
Document Digitization. This involves digitizing, analyzing, and classifying large volumes of documents, including long-term archives. It requires strong OCR capabilities, structured entity extraction, topic classification, deduplication, and maintaining legal validity.
Process unification. Linking agency workflows on a single platform to boost operational efficiency and improve decision-making quality.
Document routing optimization. Improve the routing of documents through approvals, reviews, and administrative procedures by using machine learning and AI-assisted process design.
All solutions must work with limited or no internet connectivity, prioritize open-source components, and follow planned resource allocation.
We propose a complete solution that combines reliable production components with new enterprise AI features.
Architecture Overview
The architecture includes:
🖥️ A software and infrastructure core for a distributed, fault-tolerant system.
🌊 A centralized data lake that integrates internal and external sources.
🔄 Multiple ingestion modes: API subscriptions, streaming and batch loads, and user-driven uploads.
🧹 Data cleansing and transformation, including ML-driven enrichment and classification.
🤖 An enterprise, agent-based AI that works with platform data and supports user-defined business scenarios.
📊 An analytics platform.
The infrastructure core is designed as a distributed, fault-tolerant system with interconnected tools that offer system administration, information security, authentication and authorization, and a low-code editor for managing data structures. The microservice architecture ensures horizontal scalability and isolates failures [1][2][3]. It includes built-in GitLab CI/CD and observability features like metrics, logs, and traces.
The centralized data lake combines internal and external sources, with distinct areas for raw data, cleaned data, and consumer data marts. Metadata is added to the data lake along with the main data stream and is utilized in the low-code editor.
ETL and ELT channels are set up using API subscriptions, event buses, streaming, batch loading, and user uploads with validation [4]. Cleaning and transformation include normalization, enrichment, deduplication, feature extraction, and ML classification and regression tasks. We focus on regularly addressing technical debt for data, models, and infrastructure [5]. Data loaders, the ML model software core, and CI/CD follow an Agile development approach within the project development lifecycle (SDLC).
Enterprise AI relies on LLM, vector data storage, and the n8n process orchestrator. Using LLM greatly reduces training time for specific business tasks due to their few-shot learning capability [6]. Few-shot learning is a method where models can perform tasks well with only a few training examples, helping them adapt to new tasks without much extra training. Currently, the n8n process orchestrator can use tools like microservice API calls and email sending. The feedback loop lets system administrators make controlled changes to business process rules using AI agents, supporting ongoing development.
The analytics subsystem offers tools for managing data representation through diagrams and graphs, report generation, dashboards, and OLAP data marts, all with artifact versioning and regulatory compliance.
Overall, the architecture follows the 12-Factor methodology, which includes configuration through the environment, immutable builds, build/release/run separation, and logging as event streams [7]. It also adheres to The Reactive Manifesto, focusing on asynchronous messaging, elasticity, resilience, and responsiveness [8].
Technologies Overview
The platform core uses a microservice architecture built on Kubernetes, Java, Spring Framework, TypeScript, Angular, Keycloak, Gradle, and Docker. It also includes a proprietary framework with a special constructor (low-code structure editor) that allows users to create and visualize data structures and integrate them into business processes without extra programming.
The data lake utilizes tools like MinIO and MongoDB for raw and intermediate data and metadata, PostgreSQL for cleaned data, Elasticsearch for log storage, and Redis for caching. Data is added to the lake through Kafka, RabbitMQ, Spring Cloud Data Flow, gRPC. A separate channel for critical data addition involves document digitization from various file formats, including graphics like PDF, JPG, and PNG.
Data cleaning and transformation are carried out using Python, PyTorch, scikit-learn, transformers, BERT, PyCaret, CatBoost, XGBoost, Optuna, and Apache Airflow for process orchestration.
Enterprise AI is built on Spring AI, using Mistral-24b and gpt-oss-20b as LLMs, llama.cpp for cloud deployment, Ollama for local deployment, Qdrant for embedding storage, Docker, and Kubernetes. Currently, working prototypes are being developed using LangGraph, n8n, and pgvector technologies for hybrid search scenarios.
The analytics platform employs Stimulsoft Reports, Stimulsoft Dashboards, JasperReports, BIRT, and proprietary developments.
Architecture Decision Records
ADR-001. Microservices on Kubernetes. We adopt a microservices architecture orchestrated by Kubernetes to achieve horizontal scalability, fault isolation, and independent releases. Services are built in Java with Spring, containerized with Docker, and delivered via GitLab CI/CD. This decision balances operational control with elasticity suitable for a single-tenant, isolated government cluster.
ADR-003. MongoDB-centered lake. We do not introduce Apache Iceberg because resources are constrained and the data is predominantly government-specific JSON. We choose MongoDB to support fast ingestion, flexible schema evolution, convenient APIs, and multi-document transactions for consistent data handling, even with many temporary artifacts. We implement our own custom Bronze/Silver/Gold layering on MongoDB and MinIO.
ADR-004. RAG as the foundational enterprise AI pattern. We standardize on Retrieval-Augmented Generation with Qdrant as the vector store to keep knowledge fresh without retraining LLMs. This pattern provides controllable grounding of responses and supports offline operation, provided that rigorous offline/online evaluations are conducted to manage hallucination risk.
ADR-005. BPMN for business orchestration and n8n for technical workflows. We use BPMN (e.g., Camunda) to orchestrate end-to-end business processes and n8n to execute technical steps, with an AI agent proposing changes that are subject to human approval. This separation maintains business traceability while allowing quick updates to integrations and automations.
ADR-006. Security for isolated government environments. We adopt a zero-trust approach with OIDC via Keycloak, encryption in transit and at rest with managed keys, data masking for PII. Trust boundaries are explicitly defined to reflect isolated clusters, and external interfaces are minimized while maintaining necessary interoperability.
ADR-007. In-house ML pipeline. We create our own model training and inference pipeline, storing intermediate datasets and features in MongoDB to save resources and maintain full control over execution, scheduling, and debugging. This approach reduces external dependencies, keeps costs predictable, and aligns with the need to self-host all components within the isolated perimeter.
ADR-008. Observability. We implement observability using Prometheus for metrics, Grafana for visualization, Zipkin for tracing, and ELK for logs, ensuring end-to-end visibility across APIs, data pipelines, and AI components. We deliberately exclude Istio because the deployment targets a single Kubernetes cluster in an isolated environment, and adding a service mesh would increase complexity without proportional operational benefit.
SLO/SLI
All monitoring is implemented with self-hosted, open-source tools and stored within an isolated environment.
Platform Availability SLO: 99.5% monthly in single-cluster isolated mode. SLI is measured using uptime probes and request success rates.
Data Ingestion SLO: 95th percentile end-to-end latency is under 30 minutes for daily batch loads and 8 hours for history batch loads (previous years data). SLIs are measured from the first byte received to when the data is stored.
AI Response SLO: 95th percentile inference latency is under 5 seconds for retrieval and generation on local models. SLI includes vector search latency and token generation rate.
Operations: Mean Time to Recovery is under 2 hours. Change failure rate is under 15%.
Enterprise AI
Let's explore how AI is used in the platform. The process starts with centralized data being ingested into the data lake, while content is indexed in vector storage at the same time. Indexing considers different source types like text, tables, and mixed documents.
Users create prompts for the LLM, specifying business scenarios and indicating which documents to consider directly in the user interface. The agent searches the vector storage, gathers relevant fragments, and generates results in the desired format.
The agent can then use system tools through MCP interfaces-calling microservice REST APIs, saving results to databases, sending email reports, starting the next process stage, or passing control to agents with different master-prompt instructions. This method allows for automatic system updates based on a thorough understanding of its state.
Currently, we use n8n as an experimental workflow automation tool. Users create n8n schemes to implement business scenarios. A specialized process orchestration agent suggests corrections to these n8n schemes (such as changing the order of steps, parameters, and routing), which are then reviewed by system administrators. This creates a managed feedback loop where the system evolves based on data analysis and AI recommendations.
Example Use Case
Consider a control example implementing a freight transportation data scenario. The test data corpus includes one thousand documents each of three types: freight transportation requests, work completion certificates, and invoices.
The business process is implemented as a BPMN scheme (request, carrier selection, transportation, work completion certificate, invoice) and can be integrated into the platform using frameworks like Camunda. Several n8n schemes with AI participation are implemented that analyze transportation requests and initiate cooperation, calculate quotes based on cargo data and current tariffs, send transportation status messages to social networks, generate transportation documents, and distribute them to counterparties. Here BPMN acts as the overall process orchestrator, while n8n executes specific technical actions that can be initiated by BPMN handlers, external actors (customer, carrier), or simply on schedule.
The business scenario agent receives a user prompt to analyze three document types and formulate a cost optimization strategy. The result should be saved as a report. The user works with this agent until obtaining an acceptable strategy.
The orchestration agent analyzes customer behavior (correspondence, work reviews), transportation incidents, efficiency of each transportation stage for specific routes, n8n integration fault tolerance, and formulates proposals for updating BPMN and n8n schemes. The result includes a text report and new versions of .bpmn and .json files.
Conclusion
We have examined current government agency challenges and proposed a comprehensive solution through creating a software platform using artificial intelligence. This platform provides comprehensive data source integration, document digitization and intelligent processing, and AI-based business process management. The architecture relies on proven distributed system design practices and supports a managed feedback loop for systematic improvement of data and process orchestration. This development creates a technological foundation for regulated digital transformation of government agencies.
References
Bass L., Clements P., Kazman R. Software Architecture in Practice. – 4th ed. – Boston: Addison-Wesley, 2021. – 624 p.
Tanenbaum A. S., Steen M. Distributed Systems: Principles and Paradigms. – 2nd ed. – Upper Saddle River, NJ: Pearson; Prentice Hall, 2007. – 765 p.
Newman S. Monolith to Microservices: Evolutionary Patterns to Transform Your Monolith. – Sebastopol, CA: O’Reilly Media, 2019. – 272 p.
Kleppmann M. Designing Data-Intensive Applications: The Big Ideas Behind Reliable, Scalable, and Maintainable Systems. – Beijing; Boston; Farnham; Sebastopol; Tokyo: O’Reilly Media, 2017. – 616 p.
Sculley D., Holt G., Golovin D., Davydov E., Phillips T., Ebner D., Chaudhary V. [et al.]. Hidden Technical Debt in Machine Learning Systems / D. Sculley [et al.] // Advances in Neural Information Processing Systems. – 2015. – Vol. 28. – P. 2503–2511.
Brown T. B., Mann B., Ryder N., Subbiah M., Kaplan J., Dhariwal P., Neelakantan A. [et al.]. Language Models are Few-Shot Learners / T. B. Brown [et al.] // Advances in Neural Information Processing Systems. – 2020. – Vol. 33. – P. 1877–1901.
Wiggins A. The Twelve-Factor App. – URL: https://12factor.net.
Bonér J., Farley D., Kuhn R., Thompson M. The Reactive Manifesto. – URL: https://www.reactivemanifesto.org.
