Conversational AI Experts in Germany

• Shipped Recare Voice Desktop from 0 to 1 in two months, including multi-language clinical documentation that auto-transcribes into structured German medical notes.
• Reduced LLM inference costs by 60–70% across Docs and Extract through prompt caching architecture.
• Built the AI workbench used by PMs/engineers for prompt experimentation and the Langfuse eval stack (10k+ traces evaluated).

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.

• TrendReel, production LLM agent and RAG system (Python, LangChain, OpenAI, Groq/Llama 3, Claude, FastAPI, Kubernetes, PostgreSQL).
• Designed and built a production multi-step LLM agent system: a script generation agent with a per-platform psychology database, 7 viral narrative frameworks, and structured quality scoring, switching between Claude and Groq backends in real time based on output metrics.
• Implemented multi-provider LLM routing (Claude primary, Groq/Llama 3 fallback) with priority-chain failover and quality-based provider switching, achieving 95% inference cost reduction while holding measurable quality thresholds.
• Built an advanced RAG-style retrieval pipeline with per-platform knowledge bases, semantic content matching, and structured output evaluation across 7 decision frameworks, directly analogous to multi-tenant context-based reasoning for enterprise document workflows.
• BrainyAI, adaptive AI learning platform (Python, LangChain, Groq Llama 3.3-70B, OpenAI, Next.js, Supabase, Redis).
• Integrated Groq Llama 3.3-70B with education-level-aware prompting, dynamically adjusting vocabulary depth, citation complexity, and reasoning style across four student proficiency tiers.
• Nexus Prime, multi-tenant SaaS platform for marketing and growth automation (25 modules, 99 backend routers, 153 frontend files).
• Built a 25-module, 99-router multi-tenant SaaS platform covering ad remix, affiliates, WhatsApp inbox, email, and cart recovery, serving four subscription tiers from $199 to $1,999 per month with integrated Stripe, Paystack, and Flutterwave billing.
• AI Video Surveillance Platform, multi-tenant edge and cloud computer vision system currently in active client pitch.
• Designed a multi-tenant AI video surveillance platform combining edge YOLO26 inference on NVIDIA Jetson Orin NX boxes with a central GKE cloud layer (Postgres, Pub/Sub, ClickHouse, R2, Keycloak) for event storage, dashboards, alerting, and multi-tenancy.

• AI-native B2B SaaS for freelancer sourcing; DACH market.*

• Enterprise partnerships across four industries: structured and closed multi-stakeholder deals with Telefónica (Telco), Emma Matratzen (Retail), Nürnberger Versicherungen and Flatex (Financial Services), Hubert Burda Media and Serviceplan Gruppe (Media).

• Revenue and growth: scaled FRATCH from €0 to €3.8M annual GMV, with ~80% of revenue sourced from founder-led direct outreach and partner relationships.

• Channel partnerships: sold FRATCH as a SaaS solution to recruiting firms (e.g., YER) — built the partner-enabled motion alongside direct enterprise sales.

• Team build: scaled FRATCH from solo founder to a team of 7 across engineering, product design, operations, and supply outreach.

• Proprietary network asset: onboarded 15,000+ freelancers as registered users — the proprietary DACH network powering FRATCH's matching.

• Built and launched FRATCH GPT (fratch.io/gpt): a production conversational AI agent. Architected the full stack — LLM orchestration, embeddings, re-ranking — with hands-on involvement in technical design and execution.

• GTM build: owned the full go-to-market stack — outbound, LinkedIn (organic + paid), content, and sales enablement.

• Automatic detection of content on various documents
• Recommendation Engine
• Dynamic Pricing

• ▶ 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.

struppilog.com – Digital health record for pets / MVP → Full Product

Design, development, and full further development of a digital health platform for pets – from my own MVP development to a fully built and production-ready platform.

Independent concept and development of the MVP Development of the full application with Flutter/Dart and Firebase Expansion of the MVP into a full digital health record with health data, findings, allergies, medications, documents, and emergency data Development of user registration, authentication, roles, data models, and secure user interactions Implementation of QR-code-based data exchange and digital interaction features Development of a multilingual, responsive web application Integration of AI-supported features and AI/agentic workflows Development and continuous improvement of product logic, UX/UI, and technical architecture Building and expanding a scalable cloud-based solution with Firebase Integration and further development of APIs and external services Use of AI-native / agentic engineering to speed up development, testing, debugging, and product iteration Independent implementation of all other features and technical extensions Continuous further development of the MVP into a full digital product

Impact: The MVP I built myself was continuously developed technically and functionally into a broad, production-ready platform – including frontend, backend, data model, authentication, UX/UI, APIs, cloud infrastructure, and ongoing product development.

• Built a custom AI chatbot for an e-commerce client using GPT-4 and LangChain with RAG, reducing customer support ticket volume by 45% and improving response accuracy to 92%.
• Designed and deployed an intelligent document processing system using LlamaIndex, Pinecone, and FastAPI for a FinTech startup, enabling semantic search across 100K+ financial documents.
• Developed multi-agent AI workflows using CrewAI and LangGraph for a marketing agency, automating lead research, content generation, and outreach — saving 20+ hours/week of manual work.
• Created AI-powered automation pipelines using n8n, Make, and Zapier integrated with CRMs (GoHighLevel, HubSpot), reducing manual data entry by 80% for a real estate firm.
• Delivered prompt engineering and LLM fine-tuning consulting for multiple clients, optimizing AI model outputs for customer support, content creation, and data extraction use cases.
• Built production-ready REST APIs with Python and FastAPI to serve AI models on AWS and GCP, handling 10K+ daily requests with 99.9% uptime.

• Creation of technical documentation for conversational AI products
• Development of user and API manuals (JSON, Markdown)
• Use of Git, Jira, Confluence for version control and task tracking
• Collaboration with engineers and PMs for accurate content

• Leading initiatives to improve the retrieval and representation of Samsung support content in AI-generated answers, including authoring editorial standards and educating stakeholders on representation risk.
• Developing initial frameworks to measure and evaluate brand representation in AI-generated answers (LLMs, Google AI environments), including prompt baselines and scoring models.

Consulting and prototype work on AI voice and multilingual agent systems, using AI-assisted delivery across realtime translation prototypes, call-centre automation, and voice-to-voice consultation workflows.

• Built and advised on AI voice / agentic conversation prototypes, including realtime translation and consumer-facing consultation experiences.
• Worked across call-centre automation, voice UX, product architecture, implementation tradeoffs, and prototype development.

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.

• Development, implementation, and optimization of a phone AI agent for handling existing and new customers as a freelancer

• Developing global ACM Copilot services and shaping their evolution on a global scale
• Evangelizing the benefits and capabilities of Copilot to internal and external audiences
• Onboarding, mentoring, and training new ACM colleagues in services and methodologies
• Acting as point of contact for internal Copilot proof of concept adoption and change management
• Leading Copilot adoption strategy, communicating benefits, usage and ethical considerations
• Organizing events and workshops and fostering an interactive community to share updates, findings and best practices