Applied AI systems
LLM applications, RAG, agentic systems, document intelligence, and enterprise knowledge tools designed around real workflows.
01 / Expertise
From an ambiguous problem to a system people rely on.
My work sits between technology, product, and business. The aim is not an isolated model or impressive demo, but a useful system that holds up in the real world.
Data products
Statistical modelling, predictive maintenance, anomaly detection, IoT, and large-scale time-series analytics.
Architecture & leadership
Technical strategy, software architecture, multidisciplinary teams, and the decisions that move products safely into production.
02 / Selected work
Systems built for real business processes.
A selection of problem areas I have worked on and led—from early research through production delivery.
01Documents / LLMs / Search
AI-powered contract intelligence
Systems that extract structured information, search clauses semantically, compare contract versions and supplements, and make complex contract knowledge usable.
02Agents / Tools / Workflows
Enterprise AI agents
Agentic systems that select data sources, analyze documents, call APIs and algorithms, generate visualizations, and complete multi-step tasks.
03Extraction / Validation / Automation
Intelligent document processing
Production workflows that classify, understand, extract, validate, and compare information across contracts, invoices, reports, and business documents.
04Time series / Anomalies / Operations
Predictive maintenance & IoT
Data Science systems that use high-volume operational data to detect anomalies, anticipate failures, and support better decisions in the field.
03 / Delivery
One continuous line from business problem to production.
- 01
Understand
Frame the business problem and the decision the system needs to improve.
- 02
Design
Choose an approach, architecture, and delivery path that fit the real constraints.
- 03
Build
Prototype, test, implement, and connect the system to the surrounding product.
- 04
Operate
Deploy, observe, improve, and keep the system reliable in production.
04 / About
Technical leadership, grounded in the work.
Since 2017, I have worked across Data Science, AI, and software engineering—progressing from hands-on development into company-wide technology leadership at eMentalist.
As CTO, I lead multidisciplinary teams, shape architecture and technical strategy, and work directly with customers from early workshops through delivery and rollout. I remain closely involved in the technology: evaluating AI approaches, designing systems, reviewing implementation, testing assumptions, and solving difficult production problems.
My work has spanned real estate, industrial IoT, energy, the public sector, hospitality, communications, and document-intensive enterprise workflows. The recurring challenge is learning a domain quickly and translating it into a system that works in practice.
Generative AI & LLMsRAGAI agentsStatistical modellingDocument intelligenceIoT & time seriesEnterprise softwareProduction architecture
05 / Connect
Have a hard AI or data problem?
If you are building an AI or data product, modernizing a document-heavy workflow, or looking for technical leadership, I’d be glad to hear what you are working on.
contact@patrickkiss.com