Every functioning economy depends on knowing, with certainty, who owns, uses
and holds rights to land.
and holds rights to land. When that information is inaccurate, incomplete or
unavailable, the consequences are serious: disputed titles, stalled transactions,
lost public revenue and eroded trust in government. The Land Administration
Company Inc. (LAC), based in Vancouver, exists to prevent those failures. It
provides mission-critical software and specialist services that help governments
and large institutions modernise and safeguard the systems they use to manage
land, titles, parcels and regulatory processes.
At the heart of LAC’s offering is LAVA, its proprietary cloud-native platform. LAVA
serves as the authoritative system of record for land governance, bringing together
land titles, cadastral and spatial data, legal instruments, regulatory compliance
and operational workflows in one secure, highly reliable environment. It supports
the full scope of land registry and cadastral operations, including first registration
and digital survey plan submission. For clients, this means better data quality,
faster approvals, lower administrative risk, real-time decision-making and
complete visibility and control over their land information.
LACI is the research, development and global AI engineering hub for all future LAC
products. Based in Dublin, we work closely with academic and state level
research teams to advance our mission.
About us
LACI is building new products that put applied AI at their core. We’re a small, senior team that values deep technical ownership, rigorous engineering, and building things that work reliably in the real world. We’ll share more about what we’re building during the interview process.
The role
We’re looking for a Data Engineer to turn complex, expert-authored documents into reliable, retrievable knowledge that our AI systems can reason over and cite with precision. You’ll own the pipeline from raw documents to well-structured, well-tagged knowledge: extraction, chunking, retrieval, metadata, and provenance. You’ll work closely with our Agent Systems and AI Platform & Runtime leads, our Lead Architect, and domain experts, and your work will directly determine how accurate and trustworthy our products are. This is a hands-on role with real ownership over a core part of our system.
What you’ll do
- Design, build, and operate ingestion pipelines that turn PDFs, scanned documents, and complex tables into clean, structured data.
- Choose and combine the right extraction approaches, including OCR, layout analysis, table extraction, and LLM-assisted parsing, and know when each one
- Develop chunking and retrieval strategies suited to long, structured, cross-referenced documents, and measure their impact on downstream quality.
- Design metadata and tagging schemes that capture a document’s source, version, scope, and applicability, so the right knowledge is retrieved in the right context.
- Build fine-grained provenance from source documents through to retrieved passages, so our AI systems can cite exactly where an answer comes from.
- Create evaluation datasets and metrics for extraction accuracy and retrieval quality, and use them to make evidence-based improvements.
- Work with domain experts to understand how their documents are structured and used, and turn that understanding into reliable data models.
- Handle document change over time, including new versions, amendments, and superseded content, without losing traceability.
- Contribute to our AI-first engineering culture, including how the team uses AI-assisted development tools, reviews code, and shares best practice.
What you’ll bring
- Solid software or data engineering experience, typically [5+] years, with strong production experience in Python.
- Hands-on experience extracting structured information from messy real-world documents, including PDFs, scanned material, and tables.
- Practical experience building retrieval-augmented generation (RAG) or search systems in production, including chunking, embeddings, and hybrid retrieval
- Experience designing data models, schemas, and metadata for complex or
- A rigorous, evidence-driven approach: you measure extraction and retrieval quality rather than eyeballing it.
- Care for data quality, lineage, and traceability.
- The ability to work closely with non-technical domain experts and translate their knowledge into structure.
- Clear communication with both technical and non-technical colleagues.
- Daily, fluent use of AI-assisted development tools in your own engineering work.
Nice to have
- Experience with document AI and OCR tooling, layout-aware models, or vision-
- Experience with vector databases, search engines, and reranking.
- Experience building citation, grounding, or provenance systems for LLM outputs.
- Experience working with long, formal, or highly structured documents such as standards, specifications, or regulated content.
- Experience with knowledge graphs or entity and relationship extraction.
- Experience in an early-stage or small-team environment.
- Ownership: You’ll own a core part of our system, and its quality will shape the quality of everything built on top of it.
- Challenging work: You’ll solve hard document and retrieval problems where precision really matters.
- AI-first culture: We use AI in what we build and in how we build it.
- Direct access: You’ll work closely with leadership and domain experts in a team with minimal process.
- Salary and equity: A competitive base salary plus a meaningful equity package, so you share directly in our success.
- Pension: Up to 8% employer-matched pension contribution to help you build for your future.
- Healthcare: Comprehensive health insurance contribution for peace of mind.
- Learning & Development: A dedicated annual budget for your professional growth, courses, and certifications
- Home Office Allowance: A one-off budget to help you build an ergonomic and
- Bike to Work Scheme: A tax-efficient cycle-to-work scheme to support greener,
The Land Administration Company Ireland is an equal opportunity employer. Applicants considered without regard to race, colour, religion, age, gender, marital status,