domain expertise: vocabulary-derived ownership patterns

analysis of unique vocabulary per user reveals distinct domain territories.

domain ownership matrix

domain	primary owner	evidence (unique vocab)	secondary	success rate
minecraft/fabric modding	verbose_explorer	lwjgl, netty, mixins, fabricmc, isxander, knot	—	n/a (personal)
storage engine	concise_commander	storage_optimizer, data_reorg, blocks, simd, fuzz, batch	—	84%
data visualization	concise_commander	column, canvas, chart, sort, rows, aggregation	steady_navigator	85%
query systems	concise_commander	groupby, datasets, queries, benchmark	—	70%
observability/otel	steady_navigator	opentelemetry, otel, spans, traces, attributes	concise_commander (spans)	68%
build tooling	steady_navigator	vite, pnpm, gzip, nitro, ssr, mjs	—	63%
ai/agent tooling	steady_navigator	evals, eval, oracle, apl, tool, agent	verbose_explorer	68%
devtools/amp skills	verbose_explorer	amp, typecheck, debug, patterns, notes_repo	—	n/a
react internals	verbose_explorer	react-dom, renderwithhooks, performunitofwork, beginwork	—	59%
infrastructure	patient_pathfinder	eks, prometheus, liveness probe	—	63%
streaming/sessions	precision_pilot	streams, durable, sessions, sse	—	82%
observability features	feature_lead	search_modal, analytics_service, kubernetes fields	—	45% handoff

vocabulary fingerprints

concise_commander: the data systems engineer

signature terms: pkg, column, query_engine, storage_optimizer, data_reorg, simd, benchmark, fuzz, groupby

owns the hot path. vocabulary skews toward:

columnar storage internals (blocks, rows, sort)
performance optimization (simd, benchmark, batch)
query layer (aggregation, groupby, datasets)

domain depth: deepest vocabulary density in storage-engine and query-data. terms like data_reorg and storage_optimizer don’t appear in any other user’s corpus.

steady_navigator: the platform engineer

signature terms: opentelemetry, otel, spans, traces, vite, nitro, ssr, evals, apl

straddles two territories:

observability instrumentation — otel integration, trace semantics
build/frontend platform — vite, ssr, nitro bundling

domain depth: sole owner of otel vocabulary. also primary ai-tooling contributor (evals, oracle).

verbose_explorer: the polyglot meta-worker

signature terms: minecraft, lwjgl, netty, mixins, react-dom, renderwithhooks, amp, typecheck, patterns

two distinct territories:

minecraft modding — fabric ecosystem, low-level java (lwjgl, netty)
amp tooling — skills, agents, workflow infrastructure

domain quirk: only user with react internals vocabulary (fiber, hooks implementation details). suggests debugging react at framework level, not just using it.

patient_pathfinder: the infra operator

signature terms: prometheus, eks, eu, liveness probe, readiness probe, gateway

clean operational vocabulary. no overlap with application-layer terms. pure platform ops.

feature_lead: the feature integrator

signature terms: search_modal, analytics_service, kubernetes fields, otel fields, deletion service

vocabulary centers on specific feature areas (search_modal, analytics_service). heavy on data modeling terms (fields, datasets). 45% handoff rate suggests spec-and-delegate pattern.

precision_pilot: the architect

signature terms: streams, durable, sessions, sse, timeline, migration

streaming and state management specialist. vocabulary is architectural — more about system design than implementation details.

cross-domain overlap

                    concise_commander     steady_navigator       verbose_explorer
storage-engine        ████████   -         -
data-viz              ████████   ████      -  
query-data            ████████   ██        -
observability         ██         ████████  -
build-tooling         -          ████████  ██
ai-tooling            ██         ████████  ████
minecraft/modding     -          -         ████████
react-internals       -          -         ████████
amp-skills            -          ██        ████████

vocabulary collision zones

canvas/chart — concise_commander (data layer) + steady_navigator (ui layer). both active, different depth.
oracle/ai — steady_navigator (primary), concise_commander (user). steady_navigator builds, concise_commander uses.
span — appears in both concise_commander (data structure) and steady_navigator (otel). different semantic contexts.

insights

exclusive domains (single owner)

storage internals: concise_commander. no competition. storage_optimizer, data_reorg = sole territory.
minecraft/fabric: verbose_explorer. entirely personal domain.
infrastructure: patient_pathfinder. kubernetes/prometheus vocabulary isolated.
streaming arch: precision_pilot. durable, sse not in others’ vocabulary.

contested domains

data visualization: concise_commander + steady_navigator both active. concise_commander owns data layer (rows, columns, aggregation), steady_navigator owns render layer (canvas component, chart styling).
ai tooling: steady_navigator primary builder, verbose_explorer secondary (skills/agents focus).

vocabulary as competency signal

unique term count doesn’t equal expertise depth. concise_commander’s vocabulary (18k terms) covers fewer domains but with higher density per domain. verbose_explorer’s vocabulary (21k terms) spreads across more domains with less density each.

user	domains	depth per domain
concise_commander	4	very high
steady_navigator	5	high
verbose_explorer	6	moderate
precision_pilot	2	very high

recommendations

route storage-engine work to concise_commander — vocabulary analysis confirms deep ownership
route otel/instrumentation to steady_navigator — sole owner of observability vocabulary
route platform infrastructure to patient_pathfinder — clean domain isolation
verbose_explorer for meta-tooling — amp skills, agent infrastructure, but not core product features
precision_pilot for streaming architecture — high resolution rate (82%) + deep domain vocab

generated by larry_riverbell | domain expertise analysis