Production AI systems designed for reliability.
Di Lucaro is a Stockholm-based consultancy serving teams worldwide. We pair applied AI engineering with senior systems architecture to design and ship AI systems for organizations that can't afford for them to fail.
Where we're useful.
We take on engagements where AI reliability matters — regulated industries, internal tools that have to stay correct, and products where “mostly works” is a business problem.
Enterprise RAG
01Hybrid retrieval, reranking, and evaluation harnesses for AI search that survives real users and regulated environments.
Document intelligence
02Pipelines that extract structured data from invoices, contracts, and technical documentation — with audit trails, not just outputs.
Agentic workflows
03Autonomous agents with deterministic guardrails, confidence routing, and human-in-the-loop review where it matters.
Systems architecture
04Senior systems and infrastructure design for the platform the AI runs on — distributed systems, integrations, and long-running services.
Reliability audits
05Independent review of an existing AI system: retrieval quality, prompts, evals, observability, and the failure modes you haven't seen yet.
AI visibility & content
06Retrieval-aware content architectures, thought-leadership programs, and LinkedIn strategy — how a brand surfaces inside AI answers and the writing that gets it there.
Production AI systems
- 01
Enterprise RAG, on-prem
Vermiculus Financial TechnologyHybrid retrieval (BM25 + pgvector), RRF, cross-encoder reranking. Confluence + Jira + GitHub unified. −67% time-to-information.
RAGOn-prem - 02
AI platform engineering
Walma AIMulti-tenant AI platform across four codebases — chat, ingestion, no-code pipeline builder, and serverless processing. Rebuilt the municipal invoice auditing workflow from 1/11 reliable to 11/11.
AI platformAzure
AI visibility, content & commerce
- 01
Content strategy & thought leadership
Walma AIBuilt Walma's content presence from the ground up. Over 46 days: +5,766% impressions, +1,050% followers, search appearances ~39/week → 800–1,000/week. Fully organic.
ContentStrategy - 02
Founder advisory · AI visibility
Independent clientLLM visibility strategy and analytics dashboards for a founder-led company — designing how the brand surfaces inside AI answers and measuring the result.
GEOStrategy - 03
Direct-to-consumer storefront
Stockholm independent brandEnd-to-end direct-to-consumer storefront for an independent Stockholm retailer — brand expression, product catalog, and operational integration.
CommerceWeb
How we think about reliable AI.
Reliability patterns from firmware
01State machines, circuit breakers, watchdog timers, checkpointing. AI systems need the same discipline as embedded code.
Evaluation-first design
02If you can't measure it, you can't ship it. We start with the eval that defines success — not the prompt that looks good in a demo.
Retrieval quality > model size
03Hybrid search, re-ranking, and good chunking beat bigger models. Most production wins come from the data pipeline.
Compound reliability
04A 95% step in a 5-step pipeline is 77% end-to-end. We design with the chain in mind, not the demo.
Two senior engineers. Stockholm-based, working globally.
Applied AI engineering paired with senior systems architecture. Manning Publications technical reviewers, financial markets infrastructure experience, and a firmware-trained reliability mindset.
Meet the teamCommon questions
- What does Di Lucaro do?
- Di Lucaro is a consultancy that designs and builds production AI systems — enterprise RAG, document intelligence pipelines, agentic workflows — and the systems architecture and AI visibility work around them. We focus on reliability: AI that keeps working after the demo.
- Where is Di Lucaro based, and do you work remotely?
- We are based in Stockholm, Sweden, and work remotely with clients worldwide. Engagements are remote-first, and we work in English, Spanish, Swedish, and Mandarin.
- What kind of AI systems do you build?
- Retrieval-augmented generation (RAG) systems, multi-tenant AI platforms, document intelligence and extraction pipelines, autonomous agents with guardrails and human-in-the-loop review, and evaluation infrastructure. We have shipped these for regulated finance and public-sector environments, including on-premises deployments.
- Do you work with regulated industries?
- Yes. We have built AI systems for financial technology and public-sector clients, including fully offline on-premises deployments and workflows that audit how public money is spent. Reliability, auditability, and data sovereignty are core to how we work.
- How do engagements work?
- We take on AI architecture, build, and audit engagements. Email us with what you are building, where it lives today, and what reliability looks like for you. We usually reply within two business days.