About Projects AI Work GitHub ↗

London, UK · Digital Humanities · Research Software Engineering

Yongjia Yuan 袁永甲

I build computational infrastructure for the study of ancient Christian texts — OCR/HTR pipelines for multilingual manuscripts, verifiable AI citation systems, and morphological-analysis tooling across Greek, Syriac, Latin, and Coptic. PhD in the Study of Religions (SOAS, 2026); 350,000+ words of patristic translation.

01

Manuscripts → Data

Kraken- and Tesseract-based OCR/HTR pipelines turning Greek, Syriac, Coptic, Armenian, and Geʽez manuscripts — print and medieval hands — into searchable, benchmark-gated corpora.

02

Verifiable AI

LLM systems that cite or stay silent: SHA-256 hashes and Merkle proofs bind every generated claim to a specific source passage, so hallucination fails loudly.

03

East of Rome

Research and translation where Christianity met China — the Church of the East and the Jingjiao sites of Turfan — carried by 350,000+ words of patristic translation from Syriac and Greek.

Selected Work

All projects →

CiteAgent

citeagent

AI research agent with cryptographic citation verification

A hybrid Rust + Python system that reads research materials, indexes them into a Merkle-verified knowledge base, and answers questions with deterministic, trace-bound citations: every claim maps to a specific text passage verified by SHA-256 with a full Merkle proof. Seven deterministic agents form the pipeline — ingestion, indexing, query planning, BM25 retrieval, clarification, generation, integrity verification — with a Rust (ratatui) TUI for interactive sessions. Published on PyPI as cite-extractor.

RustPythonMerkleBM25LLMPyPI
github/Areopaguaworkshop/citeagent ↗

TACF

tacf

Translation Assistant of Church Fathers — multilingual morphology platform

Full-stack application (FastAPI + Tauri + Keycloak OIDC) for Greek, Syriac, Latin, and Coptic morphological analysis and lemma lookup. SQLite-backed accentless Greek morphology index, real-time dictionary cards drawing on LSJ, Payne Smith, Lampe, and Logeion, LLM fallback for disambiguation, document upload with OCR decision routing, and Redis rate limiting behind a Caddy reverse proxy.

FastAPITauriSQLiteRedisKeycloak
github/Areopaguaworkshop/tacf ↗

msocr

msocr

Multilingual manuscript OCR/HTR toolkit

Route-aware OCR/HTR engine integrating Kraken and Tesseract for Greek, Latin, Syriac (Estrangela, Serto, East), Coptic, Armenian, and Geʽez. CER/WER benchmark reporting, Kraken model-training pipeline, FastAPI service APIs, Gradio demo UI, and RunPod training-job submission. The primary Latin print model (CATMuS-Print Large) reaches 98.56% accuracy at CER 1.44%; handwritten support covers CATMuS Medieval, 8th–15th-century manuscripts.

KrakenTesseractDockerRunPodFastAPI
github/Areopaguaworkshop/msocr ↗

Wenbi

wenbi

Multimedia → structured Markdown with AI rewriting

CLI tool converting video, audio, and URLs into VTT/Markdown, then rewriting or translating: transcription → academic rewrite, bilingual translation (DeepL first, LLM fallback), PPT slide-plus-speech combination, and batch directory processing. Configurable via CLI, Python API, or YAML. Runs in production as the transcription layer of an 80+ lecture video pipeline. Published on PyPI.

PythonWhisperDeepLPyPI
github/Areopaguaworkshop/wenbi ↗

Toolbox

Languages

PythonRustTypeScriptSQL

AI / LLM

OllamaDSPyGeminiClaude APIWhisperspaCyStanzaCLTK

OCR / HTR

KrakeneScriptoriumTesseractOCRmyPDFPyMuPDFTranskribus

Web & Infrastructure

FastAPITauriDockerKeycloakRedisCaddyRunPodGitHub Actions

Data

SQLiteSQLAlchemyTEI/XMLCSL-JSONBM25

Correspondence

yyjfwoaini@gmail.com

London, UK · open to DH / GLAM / RSE collaboration