SKF's first customer-facing GenAI assistant
A multi-agent assistant built for L1/L2 technical questions, using SKF product knowledge and retrieval.

FIELD NOTES HUMAN CURIOSITY. INDUSTRIAL IMPACT.
An engineer's mind.
A different point of view.
I build AI that earns its place in industry,
and lead the people who make it useful.
THE PRACTICEApplied AI · Industrial systems · People
THE PRESENTAI Development Manager · SKF
SCROLL TO DISCOVER01 / A LOOK INSIDE
A little window into how I think about AI.
Pick a question. Follow the source. See what holds up.
A guided example using this site's published case notes.
Ready to follow the source
Ready to follow the source
Ready to follow the source
The answer belongs here. The evidence comes first.
Choose a question, then trace its evidence.
CURATED EXAMPLE / NO LIVE INFERENCE02 / Work & impact
Customer-facing products, engineering tools in use, and outcomes reported by users.
A multi-agent assistant built for L1/L2 technical questions, using SKF product knowledge and retrieval.
AI recommendation and optimization deployed in engineering tools, using customer requirements to guide product selection.
Set the direction for AI in R&D and implemented a shared intake funnel for all AI initiatives within SKF, taking ideas into discovery and prioritization.
Shaping AI enablement across software development and embedded AI, connecting practical guidance, team capabilities, and adoption priorities. Practical software-development guidance delivered; embedded AI enablement remains planned.
03 / Selected work
One connected practice: set direction, test promising ideas, deliver useful products, and run AI where it belongs.
GenAI + decision support · Delivered
Delivered SKF's first customer-facing multi-agent technical assistant and recommendation/optimization in engineering software. Applied work also includes RAG monitoring and evaluation. Current Product Assistant support is coordinated with AI Products, which retains product ownership.
↳ Customer-facing GenAI and recommendation/optimization systems delivered for real SKF workflows.
04 / The route here
From mechanical systems in Italy to industrial AI leadership in the Netherlands. The curiosity stayed the same; the scale changed.
Turin, Italy
Mechanical engineering research at Politecnico di Torino and SKF GmbH.
Utrecht, Netherlands
Optimization, forecasting, and analytics for real engineering decisions.
SKF · Utrecht
Production GenAI, customer-facing assistants, and team delivery.
SKF · Utrecht
Leading four AI specialists and keeping delivery close to its users.
05 / On the record
Built outside work
Ideas I build and explore in my own time.
Turn a spare phone into a foreground monitor. Describe what matters, analyze activity clips with AI, and receive matching alerts through Telegram.
Explore Ping Watch07 / Notes from the work
LinkedIn posts and longer essays, in one place.
FIELD NOTE / 01A practical argument for AI leadership that can reject overbuilt solutions and protect teams from waste.
FIELD NOTE / 02A look at abstention-aware evaluation and information budgets as practical controls for hallucination risk.
FIELD NOTE / 03A grounded explanation of sampling, implementation details, and why repeat prompts can still drift.