Latest notes
Boring ML on Fabric still beats LLM-everything for the Mittelstand
A decision tree, MLflow, and one Fabric REST endpoint outprice GPT-class scoring for SMB churn, demand, and lead-scoring jobs. The talk demo proves it.
Dataverse is the agent backend most M365 shops already own
For SMBs already on Microsoft 365, the agent backend is a procurement question, not an engineering one. Dataverse ships row-level security a bolt-on vector DB cannot match.
Production RL is finally cheap enough to close the agent loop
Reinforcement learning on production agents used to be a research budget. At a hundred dollars an hour of training, it now sits next to your CI bill.
Architecting
Transformations
I am a data-driven transformation expert with extensive experience across ecommerce, energy, and financial services.
My expertise lies in building scalable data architectures, automating decision-making with AI, and delivering significant efficiency gains through robust engineering and automation.
Core Stack
- Python / SQL
- AWS / GCP / Azure
- Spark / Kafka
- LangChain / CrewAI
- Snowflake / BigQuery
- Docker / K8s
Technical Stack
Data Science
- Python & SQL
- Apache Spark
- Pandas / NumPy
- Apache Airflow
- Snowflake
- Databricks
- dbt
ML & AI Architecture
- LangChain / LangGraph
- PyTorch / TensorFlow
- OpenAI / Anthropic API
- Vector Databases (Pinecone)
- RAG Architectures
- Hugging Face Transformers
Quant & Analytics
- Time-Series Forecasting
- Risk Modeling
- Monte Carlo Simulation
- Financial Analysis
- Algorithmic Trading Strategies
- Python / R Integration
Production Systems
ML-Based Campaign Automation
Built a scalable KPI dashboard suite and data infrastructure for high-volume marketing operations. The system visualizes real-time performance metrics across multiple retail partners, drastically reducing manual reporting overhead. Deployed classification and clustering models to optimize campaign efficiency and throughput.
- Python
- AWS
- PostgreSQL
- Plotly Dash
AI Workflow Orchestrator
State-based orchestration system for AI workflows supporting high-concurrency executions with production-grade reliability. Coordinates multi-agent systems and LLM chains with sub-second state transitions.
Contextual RAG System
Financial research system processing a large-scale document corpus. Engineered advanced RAG architecture combining contextual understanding with financial data retrieval.
Forecast Modeling & Analytics
Enterprise-level dashboards and time-series forecasting models in Power BI/Azure. Improved targeting accuracy and retention forecast precision, enabling data-driven customer segmentation.
Recent articles
Mittelstand AI runs on a Mac Studio, not someone else's cloud
The typical Mittelstand AI conversation hits two walls at once. The Data Protection Officer blocks cloud LLMs, the CFO blocks the monthly bill. One workstation answers both.
Agent Skills are the cheapest ALM your team will ever buy
Most small teams cannot afford an ALM platform or the process consultants that come with it. A folder of markdown skills now does the job for free.
Pick the durable runtime before your second agent ships
Your first agent survives on a Python loop and a cron job. The second one will not, and the rewrite afterwards is the bill you wanted to avoid.
Initialize Contact
Open for collaboration on AI infrastructure, data engineering pipelines, or strategic transformation projects.
Send Message