Pydantic Expert in Germany

Built Python-based AI microservices and integrations for an AEC/VR Unity-based SaaS app, focusing on LLM/VLM capabilities, retrieval-backed systems, RESTful APIs, containerized deployment, and an automation microservice for the CAD-to-Unity pipeline.

• Developed a custom Hybrid A* based algorithm in C# to simulate hospital scenarios and detect early-stage design conflicts from collision/spatial data and generate structured reports.
• Solved and automated the time-consuming problem of converting CAD files to usable Unity environments with a custom-engineered and real-time pipeline using a ZeroMQ-based communication layer to distribute workloads across multiple processes and achieve real-time performance.
• Built a Dockerized FastAPI pipeline for CAD-to-Unity automation, combining vision-based object matching, image embeddings, and precomputed metadata to automatically map CAD objects to Unity behavior scripts, assign properties, and reduce repeated AI inference calls.
• Created documentation and examples to help technical users understand, configure, and extend the AI automation pipeline.

Conversational AI assistant for cloud infrastructure and security queries

• Designed FastAPI backend with multi-turn conversation handler, token budgeting, and context window management.
• Integrated Amazon Bedrock (Claude 3 Sonnet/Haiku); built RAG pipeline with MongoDB chat history and semantic search.
• Implemented Factory Pattern for modular LLM provider switching; reduced model onboarding effort by 60%.
• Reduced LLM inference cost by 35% through model tiering (Haiku vs Sonnet) and prompt/entity consolidation.

Tech: Python, FastAPI, Amazon Bedrock, MongoDB, Streamlit, Pydantic.

agentic-engineer.online is my publicly testable live demo and at the same time the platform where I show my work. Originally created as a recruitment trial task, I have since continued to run it as my own demo, learning, and product project — on a Hetzner VPS behind a Cloudflare tunnel, through a multi-stage AI-orchestrated deploy pipeline with snapshot rollback. If a deploy step breaks, the system falls back to the last clean snapshot, the script is adjusted, the test repeated — empirical, test-driven, without hand tuning.

• Technically behind it: Python and FastAPI, an OpenRouter model cascade, SQLite persistence, and Cloudflare edge tuning.
• I am the developer and the strictest customer of my own AI work in one person — what started as a prototype has become a tool I use every day and against which I test my own products.

• Built and fine-tuned CNN and RNN architectures using transfer learning for real-world classification tasks — core deep learning skills applicable to BMW's multimodal LLM and GenAI vehicle function development.
• Implemented Dropout, Batch Normalisation, and Early Stopping across deep learning experiments; evaluated rigorously using precision, recall, F1-score, and confusion matrices for production-grade reliability.
• Developed an AI-powered attendance management system using LBPH facial recognition, deployed via Flask web interface with real-time SMS notifications — demonstrating end-to-end AI product delivery for real users.
• Collaborated across cross-functional teams to deliver scalable, documented ML pipelines designed for reproducibility — matching BMW's interdisciplinary team and research environment.
• Integrated AI tooling directly into the development workflow from design through to testing, maintaining high velocity without compromising correctness.

• ▶ Built real-time voice assistant (STT → LLM → TTS pipeline) benchmarking and evaluating multiple STT providers including faster-whisper and Azure Speech. achieved sub-3s latency, Groq API (Llama 3) with multi-turn memory - directly handling edge cases in dictation, names and passcode recognition.

• Delivered ML pipelines for OCR, semantic search, and computer vision
• Integrated Azure AI Agents and GPT workflows for automation and QA
• Deployed cloud-based FastAPI services with scalable architecture
• Created integration docs and advised on LLM production readiness

• Lecturer for the GenAI Track at the Master School Institute of Technology
• Development of a full-stack AI application (React + Python/FastAPI) for automated supplier product import with intelligent column and category classification (4-layer hierarchical) including human-in-the-loop validation

End-to-end responsibility for all products - concept, architecture, development, and production operation as the sole developer; in addition, customer meetings, proposals, and marketing.

Underwriting Copilot - AI assistant for industrial insurance (in production at customer sites)

• Supports underwriters in analyzing industrial insurance submissions - in production use at an industrial insurer.
• Framework-independent RAG architecture with Hybrid Search (BM25 + pgvector) across large, mixed document sets.
• Two-stage evaluation and observability pipeline (code assertions + LLM-as-Judge) that makes answer quality, retrieval accuracy, and citation integrity measurable in a regression-safe way.

Kaiki Menu Analyzer - Data intelligence platform (in production at customer sites)

• Automatically captures and analyzes menu data from around 25,000 German restaurants.
• Scalable 7-container architecture (FastAPI, partitioned PostgreSQL, Redis/RQ) with LLM-supported extraction of structured data from PDF, HTML, and images.
• Full CI/CD pipelines (GitHub Actions), production cloud deployment, interactive dashboards (Dash).

Kaiki GEO Atlas - GEO platform (in production at customer sites)

• Measures brand visibility across five AI engines (ChatGPT, Gemini, Perplexity, Grok, Claude), each augmented with web search, orchestrated as a DAG workflow pipeline (Dispatcher → Sub-workflows → Scoring → Report) with fail isolation.
• 6-container deployment (FastAPI, Celery, Redis, PostgreSQL); LLM cost estimation, PDF audit report, rule-based cross-signal insights (no extra LLM cost).

Data Pipeline & Analytics Platform - competitive analysis in the automotive aftermarket

• Automated data pipeline with gap analysis algorithms and role-based access control; 230+ tests.
• Backend with FastAPI, PostgreSQL, SQLAlchemy.

Product development (actively in progress)

BankingGPT - AI assistant for complaint management in cooperative banking

• Security architecture at the core: no AI draft reaches the customer without human approval - the approval decision is in auditable code, not in the language model (monotonic: the model may escalate, never downgrade).
• Real agentic building blocks, each with its own boundary: the model chooses tools itself through an MCP server (read-only, allowlist, capped, fail-safe); sensitive cases are handed off via an open A2A protocol (JSON-RPC, Agent Card, message/send/tasks/get; client implemented by me) to a separate specialist agent (securities/law), which never lowers the review requirement (pinned by test).
• Evaluation-driven over ten analysis rounds; uncovered a security flaw through independent review and blind tests that nine automated runs had missed.
• Voice AI frontend, responding live: covered cases are answered in the conversation, sensitive ones escalate before generation; response latency < 7 s measured (local GPU STT/TTS).

Stack & production readiness: Python, pydantic-ai, FastAPI/Celery, PostgreSQL/pgvector, FastMCP, fasta2a, Docker; multi-tenant capable (physical vector isolation per tenant), PII encrypted, OWASP-LLM reviewed, 275 tests, CI/CD; vendor-portable (Ollama / EU Cloud Vertex).

After-Sales Assistant - agentic RAG/GraphRAG assistant on public OEM manuals (automotive after-sales)

• Genuinely agentic on LangGraph: ReAct agent with four tools and conversation memory - the model decides on its own whether to use the manual (RAG, Chroma), a knowledge graph (GraphRAG, Neo4j/Cypher - decodes warning lights), or a workshop/booking service.
• Human-in-the-Loop before the irreversible action: before every appointment booking, the graph pauses (interrupt) and gets the driver's explicit confirmation - the same approval-before-action discipline as in BankingGPT, in a different framework.
• Eval as CI gate: a three-part scorecard (RAGAS grounding + deterministic tool-routing accuracy + DeepEval safety: does the answer mention the warning first when there is a critical warning?) blocks the pipeline; provider-agnostic (OpenAI/Azure/Anthropic), FastAPI with token streaming.

Stack: Python, LangChain/LangGraph, Chroma, Neo4j, RAGAS/DeepEval, FastAPI, Docker.

• Bootstrapping a CloudOps team and building a multi-cloud provider backend for a low-code Internal Developer Platform (IDP) with env zero

• Introducing user story mapping, ADRs, milestones, and backlog management

• Designing and developing core APIs, setting up CI/CD pipelines, OpenTofu/Terraform scripts

• Representing and communicating the team with third-party stakeholders (e.g. env zero)

• (Cross-)team coaching on DevOps, software design, Terraform, Golang, and agile practices

• Architecture and implementation of a fully digitalized closing flow for managing securities contracts within the DVAG infrastructure. The platform aims for maximum user-friendliness, modular extensibility and compliant handling of sensitive data.

• Implementation of a reactive UI structure with a focus on user guidance & accessibility.

• Dynamic control of form and closing processes including validation logic.

• Reactive state management via SignalStore (signals + selective effects).

• UX optimization through adaptive components and Playwright-based UI tests.

• Backend modularization to connect existing sales and contract logic.

• API stability and DTO design according to Clean Architecture principles.

• Collaboration with domain teams to define technical contracts and service boundaries.

• Management with GitHub.

• Unit tests with Jest, E2E tests with Playwright.

• Code reviews, CI-integrated test execution, iterative refactorings.

• Ensuring high coverage and UI stability in the closing flow.

• Technologies: Angular 18, RxJS, SignalStore, HTML5, SCSS, Spring Boot, Kotlin, REST, OAuth2, Jest, Playwright, Clean Architecture.

AI-powered platform for automated UX validation and designer support

• Designed and led technical implementation of an enterprise-wide AI solution for automated UX review that improved design quality and significantly reduced manual review processes in teams
• Developed an automated UX validation tool as a Figma plugin and web application that generates test cases based on internal guidelines and reliably checks current designs for consistency and standard compliance
• Implemented an interactive designer chat based on RAG that answers questions about the current design and the company's UX guidelines, and designed the deployment architecture using containerized services
• Python, Azure OpenAI, PostgreSQL, REST API, Docker, OpenShift, Helm, CI/CD, Figma MCP, LLM, RAG, Prompt Engineering, GenAI, XAI, AI Architecture, AI Strategy

• Design of an LLM-based agent system to control the ERP software
• Development of agent workflows with LangGraph and PydanticAI
• Planning interfaces between business logic and language models
• Planning agent orchestration
• Prototype development and demonstration

Tools: Python, Pydantic, React, LangChain, LangGraph, Linux

• Frontend development and implementation of various designs (landing pages, overview pages)
• Responsible for the content and updates of the Datalyze Solutions company website
• Joint planning and development of an admin application for listing and locating real estate projects (CBRE)
• Joint planning and development of extensive solutions for automatic staff allocation in plants (Bosch)
• Backend development of various small projects with different stacks
• Tech stack: GitLab CI/CD, Elixir, Phoenix Framework, Ecto, PostgreSQL, Docker, Hetzner, Traefik, NGINX, Python, Flask, Next.js/React, Leaflet, GeoServer, TypeScript, MongoDB

From Crude to Crafted: Refining Participatory Design Data into Stakeholder-Ready Outcomes

• Architected a production Document AI platform using Retrieval Augmented Generation (RAG) over 1,500+ participatory design artefacts to answer historical project queries with grounded responses.
• Designed LLM evaluation combining RAGAS, custom evaluation metrics and human-in-the-loop (HITL) validation workflows to evaluate factual grounding, response quality, and prompt performance.
• Built a React, TypeScript, and D3.js frontend for interactive exploration of AI-generated insights.
• Implemented input layer LLM safety controls and Guardrails, including PII redaction and foul language filtering.