FiFE is the Feria de Informática Forense Especializada, the Spanish specialised digital forensics fair, and this is its ninth edition. It runs for a single day, Thursday 17 September 2026, at Espacio Ventas on Calle de Alcalá 182 in Madrid, organised by OnRetrieval Group. SentraLink will be exhibiting.
If you work in this field outside Spain, FiFE may not be on your calendar, and it is worth explaining why it should be. The format is unusually concentrated. One day, one venue, the manufacturers in the room rather than their resellers, and an audience drawn heavily from the Spanish state security forces, regional police, financial institutions, and the academic and legal community around them. Previous editions have been opened by senior figures from the Policía Nacional and the Guardia Civil, which tells you who is in the audience.
What the day is actually structured around
FiFE'26 splits into several strands. FiFE Academy carries the masterclass programme, where specialists present methodology rather than product. FiFE Labs is the demonstration track, where tools are shown running on real workflows. The expo floor is where the manufacturers sit. FiFE Radar, introduced last year and expanded for this edition, is a dedicated space for the technologies that are reshaping the field rather than the ones already established in it.
The speaker list this year spans investigators, vendors, and the CISO community, with contributions from institutions across Europe, North America, and Asia. That mix is the point of the event. Practitioners describing what breaks in real casework, in the same building as the people building the tools, on a single day short enough that nobody drifts away after the first session.
What we will be showing
We will be demonstrating the workflow SentraLink was built for: taking the evidence an agency already lawfully holds, in whatever shape it arrived, and correlating it into one graph without an analyst configuring anything first.
Concretely, that means loading a mixed case and watching the connections appear. Call detail records alongside bank statements. Chat exports alongside the photographs from the same device. Intercepted audio alongside the documents that were seized separately. A name mentioned in a message links to a face in an image, that face links to another image from a different device, a voice in one recording links to a second recording, and an account reference in a scanned statement links to the counterparty it paid. No ontology to design, no import specification to write, no link that an analyst had to draw by hand.
Two things about that demonstration usually generate the most discussion, and both are relevant to a Spanish and wider European audience. The first is that all of it runs locally. Every model, including language, vision, speech, face clustering, and voice comparison, executes inside the deployment. There is no external API call and no evidence leaving the perimeter, which is what makes the platform usable on live criminal material and deployable air-gapped where the unit requires it.
The second is traceability. Every finding points back to the exact source item that produced it, so an analyst can see why the system connected two things and a prosecutor can follow the same path months later. We would rather show that than talk about it, which is why we bring a working case rather than slides.
The conversations we are hoping to have
Events like this are most useful when the questions are specific. The ones we get the most out of tend to sound like the following.
What happens when the evidence is messy. A device extraction with a corrupted region, a scanned statement that OCRs badly, a call recording with two speakers over traffic noise, a document in a language nobody on the team reads. This is the normal condition of casework and it is where platforms separate.
How it sits alongside what you already own. SentraLink is not a replacement for your extraction tooling. It ingests what those tools produce, from any vendor, and correlates it with the material that never came off a device at all: carrier records, bank data, court-ordered takeouts, standalone documents. If you have an existing forensic stack, the interesting question is what happens at the seam.
What the compliance position looks like in your jurisdiction. Biometric modules carry different legal weight in different member states, which is why they are optional and configurable per deployment rather than baked into the core pipeline. If you have a specific constraint, bring it, because that conversation is more useful to both sides than a general one about GDPR.
Book a meeting
The day is short and the good slots go early. If you would like a focused session with our team at Espacio Ventas, get in touch beforehand and tell us what you want to see, and we will have a relevant case loaded rather than a generic one.
Email: contact@sentralink.com
Access to FiFE requires accreditation in advance through the organisers, so if you have not registered yet, do that first. Then come and find us.