- Sep 2026
Document AIExtracting data from PDF to Excel with AI: how it really works
Retyping data from PDFs into Excel is the first job AI can take away. Here is how a reliable pipeline works, what changes between digital PDFs and scans, and how to try it right now.
- Sep 2026
Embedded · IoTFrom sensor to dashboard: how I design data acquisition for an IoT device
Before AI models there is data, and good data is designed starting from the sensor. The method I use, from firmware to logging to the dashboard, learned on monitoring devices and energy-storage systems.
- Sep 2026
Embedded · PrototypeA glucose meter on Arduino: what an embedded prototype teaches
In 2020 I built a digital glucose meter prototype with an Arduino, an INA219 current sensor and an LCD. How it works, where the code was fragile, and why a prototype is not a medical device.
- Sep 2026
Embedded · AerospaceTelemetry for a CanSat: ground station, packets and checksums
In 2018 I wrote the telemetry ground station for a CanSat project for the AIAA competition: serial packets with checksums, flight instruments in JavaFX and a log file for every session. What I would do the same, and what not.
- Aug 2026
AI Automatic ExtractionAI automatic data extraction: how to get clean data out of messy documents
A practical guide to automatic data extraction from PDFs, invoices, contracts and scans — the architecture that actually works in production, and the four mistakes that sink most projects.
- Aug 2026
AI Call AgentsAI call agents: what they can really do for a business today
Voice AI agents that answer the phone, qualify leads and book appointments are now genuinely production-ready. Here is what works, what does not, and how to deploy one without damaging your brand.
- Jul 2026
Career · AIFrom aerospace to AI: what turbulence taught me about machine learning
I started in computational fluid dynamics at Sharif and ended up shipping production AI in Italy. The path is less strange than it sounds — both are about modelling systems too complex to solve exactly.
- Jul 2026
Process AutomationProcess automation is a mindset, not a tool
Every company wants "automation". Few can say which process, in what order, and how they will know it worked. Here is how I decide what to automate — across manufacturing, finance and enterprise .NET backends.
- Jun 2026
Career · TeamShipping AI across three countries and three languages
Iran, Germany, Italy. Persian, English, Italian. What building AI in three countries taught me about the part of the job that has nothing to do with code.
- Jul 2026
Document AI · .NETAutomating document verification inside an enterprise .NET backend
How I embed AI into enterprise .NET services at Studium Group — turning manual document review into an observable, reliable automated pipeline that scales.
- Jun 2026
AI · ConsultingHow AI actually improves internal company procedures (a practical guide)
A field-tested playbook for using AI to improve internal company procedures — where to start, what to automate first, and how to measure the return. Lessons from consulting for manufacturing, finance and services SMEs in Italy.
- Apr 2026
Reinforcement LearningWhy RL beats heuristics on the factory floor
A retrospective on replacing a rule-based scheduler at Dekali with an RL policy — and the silent rules nobody had written down.
- Feb 2026
Document AIDocument AI is just plumbing. The plumbing is the product.
Most production "AI" pipelines are 5% model and 95% taking heterogeneous PDFs, scans, Excel and rumours and turning them into a single source of truth.
- Dec 2025
AI in SMEsThe first 90 days of an AI engineer in an Italian SME
Translating "we should use AI" into a shipped feature people actually open. A field manual.
- Nov 2025
Graph Neural NetworksA GNN that learned our Bill of Materials
How we encoded stairlift components as a graph and used message-passing to predict process time and delivery date in under ten minutes.
- Oct 2025
LLM · RAGA RAG pipeline that actually answers from the manual
Lessons from shipping a customer-support chatbot grounded in 1,400 pages of Extrema manuals — and reducing internal response times by 60%.
- Aug 2025
EngineeringReading CAN-bus from energy storage, without losing your mind
Notes from the Voltfang internship: parsing CAN frames, decoding DBCs, automating conformity certificates.
- Jun 2025
Research · CFDAdding LEM to CANTERA — a research engineer's rite of passage
How a Linear Eddy Model went into the open-source CANTERA library — across C++, Fortran and Python — and what that taught me about shipping into other people's codebases.
- May 2025
.NET · AngularHow I expose Python models behind a .NET API consumed by Angular
The Dekali stack is .NET + Angular with Python models behind the curtain. Here is the wiring that keeps latency under 200 ms on the order list.
- Mar 2025
RNN · ProductionAn RNN that learned the rhythm of an Italian factory
Predicting which phase a job is in, and what it will cost by the time it ships, from telemetry no one had ever bothered to keep.
- Jan 2025
DNN · ConfiguratorFrom spec to BOM in 90 seconds: the configurator that closes deals
A deep neural net that turns customer technical specs into the optimal stairlift configuration, the full Bill of Materials and a delivery date the sales team can actually commit to.
- Nov 2024
Document AI · FinanceOld vs New: designing "I don't know" into a financial report
On the Adhoc SCF portfolio-analysis system, the most important UI affordance is not the number. It is the row that says "we cannot tell, please confirm".
- Sep 2024
Product · StrategyThe CRM is the moat, not the AI
Notes from a CRM modernisation in the Italian subsidised-finance sector. Why I keep recommending tiny AI features and bigger CRM investments.
- Jul 2024
Data · EthicsScraping check24.de without becoming the villain
Notes from the GreenPocket data project: how to do large-scale tariff scraping ethically, legally, and without breaking the source site.
- May 2024
AI in SMEsWhy AI PoCs die in week six, and how to ship in week eight instead
After three industrial projects and a dozen pitches, the pattern is the same. The PoC works in a notebook, dies on integration, and is buried by week ten.