Portrait of Simon C. Kemper as an early-modern scholar
Portrait of the engineer as an early‑modern scholar · oil on data, MMXXVI
{{ portraitHint }}
Senior · Machinarius · Data

Simon C.
Kemper

Data scientist & engineer turning archives into knowledge a machine can answer for.

Senior Data Engineer, Nationaal Archief Co‑Founder & Data Scientist, TextPast Knowledge Graphs · Linked Data · Neuro‑symbolic AI
In an age of generative AI

“If everything looks real,
what really is?”

When a model can fabricate any portrait, any document, any voice, resemblance stops being evidence. What still separates a record from an invention is provenance — and provenance is an engineering problem. Every clip on this page was generated from a historical source; the discipline is in being able to trace it back.

Background · a 17th‑century maritime painting, animated by AI

We get lost in hallucinations… see the Portfolio & Writings to find a trail back to true sources of provenance.

I.  About

Simon co‑founded TextPast to build RootReal — archival AI whose first promise is provenance.

RootReal answers questions about the past by reasoning over inventories, indices, and transcriptions — enriched through ontologies and linked to relevant external sources. A custom context‑augmentation pipeline grounds every answer in verified archival data, so each statement can be traced back to the record it came from.

It is informed by over a decade of work across government documentation, Linked Data, research, and system integration — from handwritten‑text recognition and georeferencing to knowledge‑graph engineering — together with doctoral research on neuro‑symbolic AI architectures and verifiable provenance.

See how it works in the RootReal presentation, Providing Provenance →

II.  Top Skills
01
Knowledge Graphs
02
Neuro‑symbolic AI
03
GraphRAG & Hybrid Retrieval
04
Context Engineering
05
Data Engineering
06
Machine Learning
07
GitOps & CI/CD
08
Archival Science
III.  The Record 2014 – present
OCT 2023 — PRESENT · APELDOORN, NL · HYBRID

Co‑Founder & Data Scientist — TextPast

Builds provenance‑first, neuro‑symbolic data systems for the heritage and public sectors — RootReal, a neuro‑symbolic collection‑management system, and Bronhouder, a semantic alignment workbench (bronhouder.nl). RootReal's polymorphic hypergraph on TypeDB is the runtime authority over 70+ source ontologies, and natural‑language questions are answered by SPARQL, Cypher, or TypeQL queries validated against SHACL shapes before execution — no hallucinated answers at the source. textpast.com

AUG 2021 — PRESENT · FULL‑TIME

Senior Data Engineer — Nationaal Archief

Developed the solution architecture and a natural‑language processing application that generates Linked Open Data from the Archives' inventories, indices, transcriptions, and external sources. A bespoke Rust production application feeds verified source data — archival records, relevant legislation, related curricula — to large‑language‑model workflows, so complex user questions are answered by pointing directly at fully traceable records.

FEB 2021 — AUG 2021

Scientific Researcher — Koninklijke Bibliotheek

Initiated research into a machine‑learning Named‑Entity application for the digitized Indies / Indonesian newspaper collections on Delpher (National Library of the Netherlands) and Opac (National Library of the Republic of Indonesia).

FEB 2020 — AUG 2021

Project Leader — Arsari Group

Researched Margono Djojohadikusumo (1894–1978) — founder of the Indonesian National Bank and a leading indigenous expert at the colonial Office of Co‑operatives. The biography draws on long‑term analysis of the Javanese aristocracy's socio‑economic position via handwritten‑text recognition of VOC, colonial, and Republican archives.

JUL 2019 — JUL 2020

Project Leader — The British Library

Led the digitization of over 21,000 pages of Javanese manuscripts from private collections in Central and East Java. Handwritten‑text recognition and probabilistic indexing produced the first‑ever text‑recognition model of Javanese script. eap.bl.uk

FEB 2019 — DEC 2019

Project Leader — Nikon Corporation

Led the digitization, data‑mining, and cataloguing of the Kartini Collection — the most extensive private map and print collection in Indonesia.

NOV 2018 — APR 2019

Photogrammetrist — DE REE Archiefsystemen

Applied photogrammetry and photography to digitize and catalogue objects in the museum of the Hadiningrat court of Yogyakarta. kratonjogja.id

MAR 2018 — MAR 2019

Project Leader — Delft University Library

Led the digitization of over 30,000 pages of colonial notary records from Javanese towns — long thought lost, traced by the team in private collections. Data‑mining revealed unique insights into the early‑twentieth‑century colonial real‑estate market. colonialarchitecture.eu

JUN 2016 — DEC 2017 · YOGYAKARTA

Digital Archivist — Atlas of Mutual Heritage

Analysed and georeferenced thousands of Dutch colonial maps and prints from the Rijksmuseum, the National Archives in The Hague, and affiliated institutions. atlasofmutualheritage.nl

JUN 2014 — DEC 2015 · ARSIP NASIONAL RI

Digital Archivist — Corts Foundation

Developed relational databases in Python to catalogue the diplomatic letters found in the daily records of the Dutch East Indies archives.

FEB 2012 — MAY 2014

Research Master Student — Leiden University

IV.  Education
2014–present

Doctoral Researcher (PhD, in progress) — University of Twente / Leiden University

A dissertation on the intersection of neuro‑symbolic AI architectures, Linked Data, semantic modelling, and verifiable provenance — how archival sources are modelled, enriched, and made answerable with graph technologies and NLP while keeping the trail back to the record intact.

Knowledge Graphs · Linked Data · Neuro‑symbolic AI · NLP
2012–14

Master of Arts (MA), Digital Humanities — Leiden University

Graduated Cum Laude.  Data Analytics · Analytical Skills

2008–11

Bachelor of Science (BSc) — Utrecht University

Methodology & Statistics, with a track in history. Graduated Cum Laude.

V.  Skills & Languages

Knowledge Graphs & Semantic Web

RDF RDFS OWL / OWL2 SPARQL SHACL SKOS / KOS RiC‑O EAD LinkML SSSOM Linked Open Data Ontology Lifecycle Taxonomy Alignment

Graph & Vector Databases

TypeDB (hypergraph) Neo4j Oxigraph Qdrant Weaviate FAISS DuckDB / DuckLake

Retrieval & AI Architectures

GraphRAG / Neuro‑symbolic RAG Hybrid Graph + Vector Retrieval Semantic Search Schema‑validated NL Querying Agentic AI Machine Learning Named‑Entity Processing Handwriting Recognition (HTR) OCR

Data Products & Engineering

Canonical Data Models Master‑data Pipelines Entity Matching & Resolution Persistent Identification Microsoft Fabric SharePoint Ingestion GIS

Engineering Practice

Rust Python SQL JavaScript Git CI/CD Automated Testing Docker GitOps Evaluation & Benchmarking

Languages

DutchNATIVE
EnglishFLUENT
Bahasa IndonesiaFLUENT
FrenchPASSIVE
GermanPASSIVE
JavanesePASSIVE
Portfolio & Writings

Provenance, Engineered

Selected work in archival data engineering — systems that turn four centuries of inventories, transcriptions, and maps into knowledge a machine can answer for, without ever losing the trail back to the source.

Selected Work
01 2021 – present

RootReal — the Context‑Augmentation Pipeline

TextPast

A bespoke Rust pipeline that grounds every language‑model answer in verified archival data — reasoning over inventories, indices, and transcriptions, enriched through ontologies and linked sources. Each statement it produces can be traced directly back to the record it came from. Provenance is not a feature bolted on at the end; it is the architecture.

Rust LLM Grounding Context Engineering Provenance
See “Providing Provenance” →
[ architecture diagram
— source → pipeline → grounded answer ]
02 2021 – present

Archival Assistent (Aa)

TextPast

A natural-language assistant that automates the creation of archival aids and answers visitors' questions about the collection — reasoning over inventories, indices, and transcriptions as Linked Open Data, with every answer pointing back to the record it came from.

Archival Aids Natural-Language Q&A Linked Open Data
03 2019 – 2020

The First Text‑Recognition Model of Javanese Script

The British Library · EAP1268

Led the digitization of 21,000+ pages of Javanese manuscripts from private collections in Central and East Java. Handwritten‑text recognition and probabilistic indexing produced the first‑ever HTR model of Javanese script.

HTR Probabilistic Indexing 21,000 pages
eap.bl.uk →
04 2018 – 2019

Colonial Notary Records, Recovered

Delft University Library

Led the digitization of 30,000+ pages of colonial notary records from Javanese towns — long thought lost, traced by the team in private collections. Data‑mining revealed unique insight into the early‑twentieth‑century colonial real‑estate market.

Digitization Data Mining 30,000 pages
colonialarchitecture.eu →
05 2016 – 2017

Georeferencing the Atlas of Mutual Heritage

Atlas of Mutual Heritage · Yogyakarta

Analysed and georeferenced thousands of Dutch colonial maps and prints from the Rijksmuseum, the National Archives in The Hague, and affiliated institutions — projecting historical cartography onto modern coordinates.

GIS Georeferencing Cartography
atlasofmutualheritage.nl →
Writings & Talks
Presentation

Providing Provenance — the RootReal architecture

How archival sources are modelled, enriched, and made answerable by a language model without losing the trail back to the record.

Read →
Thesis

Modelling the Archive (in progress)

A doctoral study of neuro‑symbolic AI architectures, Linked Data, semantic modelling, and verifiable provenance — graph technologies and NLP applied to centuries of archival sources, at the University of Twente and Leiden University.

In progress
Method

Animating the Source, Keeping the Trail

Notes on the gallery above: generating motion from a documented original, and why the provenance record — not the result — is the deliverable.

Notes