Skip to content
View Blowcake's full-sized avatar
  • Germany / NRW

Block or report Blowcake

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Blowcake/README.md

Hi there, I'm Harald Frank

AI systems, automation, and process integration.

I build local-first AI systems, automation tools, and integration layers that connect LLMs with real business processes, industrial systems, documents, and structured data.


Focus & Development Philosophy

  • Local & Hybrid AI Integration: Developing secure architectures that connect LLMs to SQL databases, local document caches, and Office suites using custom Model Context Protocol (MCP) servers.
  • Full-Stack IoT Orchestration: Coordinating complex IoT/IIoT platforms from hardware design and firmware to cloud infrastructure and mobile apps.
  • Compliance & QM/QS Automation: Engineering automated systems that parse documentation, calculate risks, and streamline quality assurance process chains.

Technical Skillset

Category Technologies & Tools
AI & LLM Systems LLM runtimes & inference stacks, local model deployment, RAG, agentic workflows, Model Context Protocol, OCR, document AI, computer vision
Languages Python, SQL, AutoIt, PHP, C/C++, JavaScript
Web & Application Development HTML, CSS, frontend, backend, REST APIs
Architecture & Systems Microservices, event-driven systems, local semantic memory
IoT & Industrial MQTT, Modbus, TCP/IP, Bluetooth LE, WLAN, CAE engineering systems

Key Developed Projects

Enterprise-grade compliance, risk, and project management suite

  • Full product compliance and risk management suite covering lifecycle evaluation, reporting, project governance, and automated declarations of conformity.
  • Analyzes Bills of Materials (BOM), supplier certificates, test reports, technical documentation, and product specifications.
  • Uses LLM-based document analysis to interpret compliance evidence and validate applied standards, directives, and regulations.
  • Supports structured decision-making across European and national product compliance requirements.

Model Context Protocol client and server stack (Tier 1 grade classification)

  • Full MCP client and server stack covering tool servers, resource providers, protocol-safe orchestration, local integrations, and structured AI-to-application bridges.
  • Includes a reusable MCP client layer used by the multi-persona agent engine to assign role-specific tools, resources, and integration capabilities.
  • Connects LLMs natively to local Office environments such as Excel and Word as well as SQL databases.
  • Enables schema discovery, FTS5 semantic searching, and structured database operations for AI agents.
  • Designed for robust local-first AI workflows with clear tool boundaries, reusable server components, and secure integration patterns.

Multi-agent coordination orchestrator

  • Multi-persona agent engine with template-based agent definitions, flexible skills, and role-specific execution profiles.
  • Supports individual LLM settings per agent, including model selection, runtime configuration, and task-specific behavior tuning.
  • Supports individual MCP capabilities per agent, allowing each agent to access only the tools, resources, and integrations required for its role.
  • Includes highly specialized agents for document analysis and validation of applied standards, directives, and regulations against current compliance requirements.
  • Manages concurrent AI agents equipped with local memory and RAG layers.
  • Implements multiple safety layers for autonomous execution, including scoped tool access, validation gates, escalation rules, and controlled action boundaries.
  • Designed for autonomous 24/7 operation with Human-in-the-Loop validation gates for critical decisions and controlled escalation.

Areas of Domain Expertise

Technical Foundation & Product Quality

  • Formal background in communication electronics and information technology, combined with long-term experience in product development, product management, quality management, compliance, and software-driven process automation.

Enterprise AI & LLM Systems

  • Designing and hosting local model architectures for secure corporate data processing.
  • Building advanced retrieval workflows and semantic project indexing over local filesystem directories.

Industrial IoT & Measurement Platforms

  • Coordinating large-scale IoT system developments spanning custom hardware, firmware, cloud endpoints, mobile app synchronization, and serial manufacturing.
  • Working with sensor protocols and radio communications including MQTT, Modbus, Bluetooth LE, and WLAN.

Automated Quality Management & Compliance

  • Automating QM/QS process chains, digital tracking workflows, and management dashboard generation.
  • Building document analysis pipelines for certificates, test reports, and compliance declarations.

CAE Process Engineering

  • Trained and supported CAE users on customer systems, including practical workflows, system usage, and process-oriented application of engineering tools.
  • Created electrical schematics and circuit documentation directly within CAE environments.
  • Analyzed plant and machinery engineering schematics to identify rationalization and software automation potential.
  • Customized CAE/CAD scripting workflows and system tools.

Popular repositories Loading

  1. Blowcake Blowcake Public

    Special repository for Harald Frank's GitHub profile README.