Production specifics.
Numbers, dates, named clients. Every engagement is 6+ months in production, not a demo.
Specifics, not promises.
Specifics. Real numbers, real dates, real client names — none of the generic promises.
- inFakt
inFakt — MCP server, AutoX, PKD AI: products that wire accounting into LLMs
- 300k+invoices / month
- 90%+OCR + NLP accuracy
- −40%data processing cost
- 30%projected efficiency gain
I lead the AI Engineering team. Three flagship products built by the team: a Model Context Protocol server (launched, with read and write tools), the AutoX automated bookkeeping system (300k+ invoices/month, 90%+ accuracy), and the PKD AI classifier for Polish business registration. Plus a modern AWS data architecture. Docs: infakt.ai.
Role: AI Engineering ManagerStack: Python · LangChain · Anthropic MCP SDK · AWS S3 / Redshift · dbt · … -
- IG Group
IG Group — KYC automation across 40+ languages: OCR + GPT-4 + multilingual NMT
- −35%account opening time
- 89%accuracy across 40+ languages
- 120h+saved / month
- 1,200+documents
Architected and deployed an end-to-end automated workflow for verification and processing of international KYC documents. OCR + GPT-4 + machine translation. Reduced onboarding by 35%, 89% accuracy across 40+ languages, 1,200+ documents processed. Full Compliance and Legal collaboration.
Role: Data Science Team LeaderStack: Python · OCR · GPT-4 · Azure OpenAI · Machine translation · … -
- IG Group
IG Group — first generative AI in a financial organization: $100k+ annual savings
- $100k+annual savings
- $40kmonthly added revenue
- $250kclient-scoring savings
- 100shours/mo eliminated
Pioneered generative AI deployment at a global trading broker: internal knowledge base chatbot, AI-powered compliance verification, machine translation for SEO/agency/client comms. LangChain + AzureOpenAI + Vertex AI. $100k+ annual savings, hundreds of manual hours/month eliminated.
Role: Senior Data Scientist / AI EngineerStack: Python · LangChain · AzureOpenAI (GPT-4) · Vertex AI (PaLM, Gemini) · Streamlit · … -
Let's talk.
30 minutes, no obligation. Tell me where your AI initiative is stuck or what you're planning — you'll leave with concrete next steps.