cognition
The Shadow of Compression
On symbolic and statistical learning as two directions toward the same fixed point
Data Scientist · Complex Systems Analyst · Independent Researcher
I’m a systems engineer applying model-based systems engineering to AI, organizational knowledge, and human decision-making.
I design modular AI workflows, information retrieval systems, evaluation frameworks, and feedback loops. My work focuses on making the knowledge, decisions, and technical infrastructure behind AI systems observable, testable, and easier to steer.
Questions I’m exploring
- How can structured disagreement lead to better decisions?
- Where does determinism help—and where does it create false confidence?
- How do institutional checks and balances make human and AI decisions trustworthy?
Recent thoughts
code
Building Is Cheap. Knowing What to Build Isn’t
Reflections on building tools at the boundary of math, software, and understanding
Selected projects
Recent work
Active · Open source
gds-core
A typed, compositional framework for specifying, simulating, and verifying complex systems.
Research · Open source
Latent Topologies
Research into the geometry of language-model representations using persistent homology and Hodge decomposition.
Active · Open source
Murmur
A local-first Linux tool for recording, transcribing, and summarizing meetings without a cloud meeting bot.
Research notes
Information Retrieval as Epistemic Architecture
A research notebook about representation, bounded attention, and the systems that make knowledge retrievable.
Unfinished by design
Work in progress
Half-built tools, research threads, and ideas I’m still testing. These may change direction, combine with something else, or quietly disappear.
Cataloguing
Practical AI Catalogue
Mapping deployable multimodal AI capabilities to real business functions, constraints, and evaluation methods.
Thinking in public
Systems Thinking Notes
Connecting software architecture, control theory, system properties, and the views we use to reason about them.
Prototype
LLM Ranker
An experiment in comparing coding models using cost, reliability, tool-use accuracy, and task-specific tradeoffs.