Research & playbooks
for shipping AI in Asia.
Frameworks we use in client engagements, plus original research on AI adoption across the markets we operate in. No hype, no rehashed Western reports.
Multi-Agent AI Systems: Enterprise Design Patterns for APAC Deployments
The first generation of enterprise AI was single-agent: one model, one task, one output. Multi-agent systems unlock compound tasks — but they introduce orchestration complexity and new failure modes. Here are the patterns that work in production.
RAG vs Fine-Tuning vs Prompting: Which Pattern Fits Your Use Case?
Three deployment patterns, three sets of trade-offs. A decision tree that picks the right one for your AI use case in under five minutes.
Building AI platform teams that ship
Recording and artifacts from our recent session with CTOs across the region.
Manufacturing AI in Greater China: Computer Vision QC, Predictive Maintenance, OEE
Three manufacturing AI use cases delivering measured value in Greater China today, with the integration patterns that make them work on the factory floor.
Choosing a Vector Database in 2026: 7 Options Compared
Seven vector database options used in production by mid-market enterprises in Asia, with cost, latency, and operational profile for each.
Agentic AI in Production: Lessons from 12 Mid-Market Deployments
Twelve production agentic AI deployments across Asia in 2024-2025, with the patterns that worked, the patterns that did not, and what is replicable.
From Pilot to Production: An MLOps Maturity Model for Mid-Market Teams
A four-stage MLOps maturity model designed for mid-market AI teams, with the practices to add at each stage and the practices to skip.
Webinar Recap: RAG in Production — What Breaks at Scale in APAC Enterprise
Sixty-one engineers and architects attended our January session on retrieval-augmented generation in enterprise production environments. The questions were sharp.
AI Cost Engineering: How to Drop Inference Costs 60% Without Losing Quality
Production LLM systems often have 50-70% of inference cost going to non-essential work. Five techniques recover most of that, with no quality loss.
Want these in your inbox?
Subscribe to the RSS feed or talk to us about a research engagement on a topic specific to your firm.