My Blog
Reflections from My Journey
Reflections, technical deep dives, and opinions from a decade at the intersection of humanitarian data, GIS, climate risk, and cash programming.
Why Disaster Loss Data Matters More Than Ever for Climate Adaptation
In Cox's Bazar, host communities pushed back against reforestation — not because they opposed it, but because their own climate losses to coastal erosion and cyclones were undocumented and therefore unfundable. Disaster loss data is now the evidentiary backbone of the entire climate adaptation architecture.
Read Article →From Forecast to Action: Operationalising Early Warning and Anticipatory Action with DELTA Resilience
A meteorological forecast tells you what is coming. Historical loss data tells you what it will do when it arrives. DELTA Resilience is the first national disaster data system designed to provide that missing link at scale — turning early warnings into impact-based, evidence-driven anticipatory action.
Read Article →Invisible Disasters, Invisible Funding: When Disaster Data Decides Who Gets Climate Finance
Every year, millions experience flash floods, prolonged drought, and slow-onset hazards that never reach the world's primary disaster databases. Their losses are real, recurring, and devastating. Because they don't show up in the data, they rarely show up in the funding either.
Read more →The 72-Hour Post Disaster Problem
The first 72 hours of a sudden-onset disaster are an information black hole. Good IM isn't about perfect data — it's about being useful under imperfect conditions.
Read more →The Case for Anticipatory Cash
We can predict most slow-onset disasters weeks in advance but still wait for them to happen before responding. Every dollar spent before a flood is worth five dollars spent after.
Read more →Measuring Joint Response for Cash Transfer Programmes — A New Way of Using Humanitarian Meta-Data
Five organisations running cash transfer programmes in the same country produce five sets of post-distribution monitoring data using five different tools. The simple question — "is our collective cash response working?" — becomes structurally unanswerable. Inter-agency PDM meta-analysis is how you answer it.
Read more →GeoAI for Humanitarians: Getting Started
GeoAI has enormous potential for humanitarian operations, but most IM officers don't know where to start. This is a practical guide.
Read more →The IM Coordination Trap
The biggest barriers to good information management in humanitarian response are not technical — they're political. Data sharing agreements and institutional distrust kill more IM initiatives than bad technology.
Read more →The Future of Humanitarian IM is Agentic
AISA and why the next generation of humanitarian information management will use AI agents, not just AI tools.
Read more →Voice Is the Future of Humanitarian Data and Evidence Generation
After a decade of building form-based reporting systems across six countries, voice AI will fundamentally reshape how the humanitarian sector generates evidence. The interface was always the bottleneck.
Read more →From Forms to Voice: The Deeper Inclusive Transition
Every number in our reporting systems started as a human observation that had to survive a form before it became actionable. Voice-to-schema AI ends that entire pipeline.
Read more →From Reporting Platforms to Voice-Powered Decision Intelligence
A field officer in Kabul told me: "By the time our data reaches Kabul, the situation has already moved." Voice AI combined with agentic AI collapses the pipeline from weeks to seconds.
Read more →Building Voice-Native Evidence Systems: From Theory to Architecture
What does a voice-native humanitarian evidence system actually look like? After building form-based platforms for a decade, here's the architecture — and why it changes everything.
Read more →Building Disaster Data Systems That Governments Can Own
A flood vulnerability analysis I designed died quietly two years after I left. The hardest lesson from a decade of building these platforms isn't technical — it's institutional.
Read more →The Evolution of National Disaster Tracking Systems: From DesInventar to DELTA Resilience
Not a software upgrade — an architectural paradigm shift from a standalone record-keeping tool to a sovereign, interoperable, AI-ready data ecosystem. Why and how the world outgrew DesInventar.
Read more →The Global Disaster-Related Statistics Framework: Why Statisticians and Disaster Managers Must Finally Speak the Same Language
Endorsed by the UN Statistical Commission in March 2026, the G-DRSF gives disaster managers and statisticians a shared vocabulary, shared standards, and a shared reason to work together.
Read more →The Data Ecosystem Maturity Assessment: A Practitioner's Guide to Diagnosing National Disaster Data Readiness
A maturity assessment is not a delay. It is the investment that ensures the system you build is the system that survives. The DEMA framework, in practice.
Read more →The Politics of Humanitarian Data Infrastructure: Who Owns the System When Everyone Walks Away?
The email I sent at 11am to 115 organisations announced the platform was suspended immediately. Afghanistan in 2025 was a stress test that revealed a system-wide architectural flaw: nobody owns continuity.
Read more →From Crisis Zones Digital Systems to Market Zones Digital Transition for Africa's Informal Economies
While working abroad over the last decade, I visited Nigeria every few months. Every visit, the same struggle — finding reliable services, navigating markets blind, and watching trust deficits hold back an entire economy from going digital. Then a realisation hit me.
Read more →Why I Build Systems, Not Dashboards
The humanitarian sector is drowning in dashboards but starving for systems. A dashboard is a view; a system is an ecosystem that changes how organizations make decisions.
Read more →Africa Will Define How the World Uses Voice AI
Africa skipped landlines for mobile. Skipped bank branches for M-Pesa. Next: skipping text-based interfaces for voice-first AI. And this time, the continent won't just adopt — it will lead.
Read more →The Voices Our Data Systems Were Built to Silence
Accountability to Affected Populations has been a humanitarian commitment for over a decade. But our data collection tools — forms, checkboxes, pre-coded categories — were never designed to listen.
Read more →Voice Infrastructure Inequality: The New Digital Divide
AI scores 80% accuracy in English. Below 55% in Yoruba, spoken by 50 million people. If voice is the future of data, voice infrastructure inequality is the future of data exclusion.
Read more →Lessons from Building Humanitarian Data Platforms Across Multiple Crisis Contexts
Multiple countries. Seven data platforms. A decade of work. Six principles emerged across all of them — and none are about technology.
Read more →