Pydantic Experts

• Claude Code subagents, MCP, Pydantic V2, pytest, bandit

• Designed and shipped 3 specialized agents that hand work down a line: a research agent writes a cited implementation spec, a coding agent builds the modular code and its tests, a review agent ranks findings by severity and applies the fixes. Each handoff is a structured document, so no stage depends on another agent's context window.

• Connected the research agent to an academic-research MCP server (Semantic Scholar, ArXiv, Hugging Face Hub, citation snowballing) so every reference traces to a tool result rather than the model. Gated commits behind ruff, mypy, pytest and bandit, required human sign-off before installs and commits, and persisted session state on disk so long runs survive a context reset.

• Designed and developed a multi-tenant SaaS platform enabling organizations to build their own knowledge bases and chat with brand-customized AI assistants (white-label approach with dynamic branding per organization).
• Implemented a scalable RAG architecture with a GPT-4o tool-use loop, hybrid semantic search, and strict tenant isolation at database and search index level.
• Built persistent, project-like chat sessions including a streaming API (SSE), multilingual support, and speech input/output (STT/TTS).
• Delivered the cloud infrastructure as Infrastructure-as-Code, fully automated per-customer CI/CD pipelines, and an onboarding process for new tenants.

Technologies used: Python, FastAPI, Pydantic (v2 noted), Next.js, React, TypeScript, Tailwind CSS, OpenAI / LLMs (GPT-4o), Azure AI Search, Cosmos DB, Azure Blob Storage, Azure Cognitive Services Speech, Azure App Service, Azure Container Registry, Retrieval-Augmented Generation (RAG), Server-Sent Events (SSE), Docker, Terraform, GitHub Actions, REST, OpenID Connect (OIDC), Multi-Tenancy

Designed and developed an Azure-based AI and data platform transforming distributed project documents into searchable, structured knowledge for an enterprise AI assistant. Combined incremental document processing, hybrid keyword/vector retrieval, knowledge graph extraction, and conversational AI for Microsoft Teams and Microsoft 365 Copilot. Implemented Python document-processing pipelines for PDF, Word, Outlook and text files; Azure Cosmos DB lifecycle management; Azure AI Search; Azure AI Foundry and Azure OpenAI; identity-aware retrieval with Microsoft Entra ID; project authorization; a TypeScript Microsoft Agents SDK agent; Pulumi infrastructure; resilient scheduling, retry and diagnostic tooling. Technology stack: Python, TypeScript, FastAPI, Pydantic, Azure AI Foundry, Azure OpenAI, Azure AI Search, Azure Cosmos DB, Blob/Table Storage, Container Apps/Jobs, Service Bus, Key Vault, Entra ID, Docker, Pulumi, pytest, Microsoft Teams and Microsoft 365 Copilot.

• Fine-tuned and evaluated alternative backbone architectures for the AFFGAN handwriting style encoder to better capture writer-specific features, benchmarking against the original baseline architecture and replacing the legacy VGG19 architecture with a custom Transformer-based encoder.
• Extended the system beyond word-level synthesis to line-level handwriting generation, integrating and adapting an existing line-generation architecture to support full-sentence output.
• Deployed the handwriting generation pipeline as a web application, enabling users to select a writing style, input custom text, and instantly generate a corresponding handwriting image.
• Integrated TrOCR into a teacher–student distillation framework to evaluate and improve readability of generated handwriting, improving OCR accuracy by 20% over the baseline.
• Boosted user-study pick-rates by 40%, demonstrating a significant leap in the perceptual quality and realism of the generated handwriting compared to the baseline model.

Tech: PyTorch, Python, GANs, CNN, ViT, TrOCR, Streamlit, CUDA.

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.

• ▶ 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.

Independent engineering practice focused on custom AI systems, production delivery, and fractional technical leadership.

Selected client engagement: Scalutions

• Role: Serve as fractional CTO and hands-on technical lead, responsible for the architecture and agentic infrastructure behind its managed B2B outbound operation.

• Product: Designed and built OutboundLoop, an agentic SDR operating system for research, qualification, personalized outreach, campaign management, human approvals, measurement, and continuous improvement.

• Scope: Own the full system lifecycle—from business processes and agent behavior to context design, model routing, integrations, evaluation, telemetry, reliability, cost control, and production operations.

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.

• Taught Python, Pandas, data analysis, visualization and Power BI to 250+ students.
• Guided practical projects from data preparation and exploratory analysis through visualization and dashboard creation.
• Explained complex analytical concepts clearly to audiences with different levels of technical experience.

Project: Tax Exemption Order Application

The client has an existing application for creating and maintaining tax exemption orders for end customers; design and implementation of a comparable application for internal employees.

• Design and implementation of microservices and the UI for the business area "tax exemption orders" using Domain Driven Design as well as Spring Boot and Angular.
• Implementation of reactive, non-reactive, and asynchronous APIs (Spring REST, WebFlux, GraphQL).
• Development of the Angular application, including state management using Signals, RxJS Observables, and subscriptions.
• Securing the API and the application using OAuth2, JWT, and OpenID Connect.
• Configuration and setup of CI/CD pipelines with Jenkins.
• Collaboration with cross-functional teams and conducting code reviews.

Environment: Java, Spring Boot, Angular 18 & 19 (standalone, signals), RxJs, Bootstrap CSS, Vitesting, OpenShift, Istio, microservices, Kafka, Dynatrace, Jenkins, GitLab, Graylog, Sonar, Oauth2, OracleDB

• Built a Python pipeline that converts lectionary web content into structured daily JSON, applying liturgical-calendar rules for accurate weekday and Sunday coverage.
• Shipped a read-only FastAPI REST API with shared Pydantic models and delivered a Kirundi-first web client for browser and mobile use.
• Owned the data layer and backend architecture, collaborating closely on system architecture and interfaces while automating refreshes with GitHub Actions and validating the ETL with pytest.
• Impact: Created a reliable, API-driven source of truth for daily Bible readings in Kirundi, enabling consistent access to previously unstructured content.

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.