{Jacob Dorrill}
ML Engineer — RAG Systems, AI Agents, and Predictive Models
I build production ML systems — RAG pipelines with pgvector, AI agents wired to real databases via MCP, gradient-boosting predictive models, and OCR extraction pipelines. Currently shipping ML features at CompuCal (Cork, Ireland) for a regulated calibration-management platform.
RAG Pipelines
Embeddings, retrieval with confidence scoring, and citation surfacing. pgvector, LangChain, FastAPI — served as a REST API or embedded widget.
AI Agents with Tool Access
Multi-agent orchestration via CrewAI. MCP servers connecting agents to SQL databases, internal APIs, and documentation repos — not toy prompt chains.
Predictive ML
Classification, regression, and anomaly detection with CatBoost, XGBoost, LightGBM, and ensembles. Benchmarked against a real baseline, not just cross-validated in a notebook.
OCR & Document Extraction
Structured JSON extraction from scanned PDFs. Azure Document Intelligence or in-house GLM-OCR depending on cost and accuracy tradeoffs.
My Journey
I started with Python scripts in early 2021 — automating crypto trades was the excuse to learn the language. Five years on, the work has settled: production ML for small teams. RAG systems, AI agents with real tool access, and predictive models for tabular and document-heavy problems. Currently shipping ML features at CompuCal, BSc at MTU Cork just wrapped.
Education
BSc Software Development, MTU Cork (completed). QQI Level 6 Advanced Software Development with distinction prior.
Current Role
Shipping ML features at CompuCal (Cork, Ireland) — RAG, OCR, predictive maintenance, and MCP-backed agents for a regulated calibration-management platform.
Focus
Production ML for small teams: retrieval systems, AI agents, and predictive modelling for tabular and document-heavy problems.