Thinking in Public
The Cognitive Data Science Lens
Each piece here applies cognitive science principles to data and AI challenges. This isn't just technical writing—it's a systematic exploration of how human cognition shapes (and should shape) the systems we build.
Canon: Foundational Frameworks
Start here. These pieces define my approach to data, AI, and system design.
Vision, Perception, and Data Viz for Decision-Making
Why great charts aren't pretty—they're perceptually efficient. Using preattentive cues, luminance over hue, and small multiples to answer one question per view.
Bees, Graphs, and Governance
How treating beehive photos as a knowledge graph revealed universal patterns for enterprise data governance. Sometimes the best systems emerge from the messiest domains.
RAG Without the Theater
Evidence-linked retrieval patterns you can defend. Skip magic prompts; use small, testable patterns that tie claims to sources.
Why Your Dashboards Fail: A Cognitive Scientist's Perspective
The 200-millisecond problem, working memory limits, and why humans don't want data—they want narratives. With examples from my fitness tracker redesign.
Data Stories: Cognition in the Wild
Real projects where cognitive principles meet messy reality.
The Choco Effect: How a Dog Transformed My Running Data
When my running data collided with dog walks, it revealed how consistency beats intensity—and how life's interruptions make better data stories than perfect tracking.
From Owl Box to Data Pipeline: A Beekeeper's Digital Journey
How 400+ unlabeled bee photos became a knowledge graph using EXIF metadata, weather APIs, and computer vision. A masterclass in finding structure in chaos.
Knowledge Cartography: Finding Lost Cousins in the Academic Family Tree
Using TransE to predict which papers should be citing each other but aren't. When your old research becomes a treasure map to hidden connections.
The Convoscope Chronicles: Building Multi-Modal AI Memory
Why comparing AI models side-by-side revealed the importance of cognitive load management in human-AI interfaces.
Building & Learning in Public
Technical journeys, lessons learned, and works in progress.
IoT Sensor Fleet: From Arduino to Analytics
Building a temperature sensor network taught me that real-time data is only valuable if humans can understand it in real-time too.
Digital Home Base Workshop Series
Teaching others to build portfolio sites revealed how information architecture principles apply to personal branding.
From Notebook to Production: A Cognitive Approach
Why the journey from Jupyter to production isn't technical—it's cognitive. Managing complexity for future-you.
Quick Takes & Observations
Shorter thoughts on data, AI, and human-system interaction.
The 7±2 Rule Everywhere
Why Miller's magic number appears in everything from Streamlit interfaces to LLM context windows.
Attention Mechanisms: Biology to Transformers
The fascinating parallels between human visual attention and transformer architecture.
Why Experts Make Terrible Interfaces
The curse of knowledge is real—and it's ruining your dashboards.