Projects
Applied AI — on top of a decade of delivery.
Two threads of work: the AI tooling I’m building now at the intersection of LLMs, data engineering, and self-serve analytics — and the foundation of shipped governance, forecasting, and team-building that came before it.
Applied AI · current focus
LLMs in production for data teams.
Conversational BI for leadership
shippedNatural-language interface over a governed LookML semantic layer so leadership can query BigQuery without writing SQL. Guardrails, citations back to source models, and an eval harness for accuracy.
Analyst Copilot in the reporting suite
shippedAI assistant embedded in our Looker workflow that drafts narratives for weekly readouts, summarizes anomalies, and answers stakeholder follow-ups in context.
Teradata → BigQuery migration assistant
in-progressLLM-assisted translator for legacy Teradata SQL into BigQuery models, paired with a deterministic diff harness that validates the new model produces identical rows before cutover.
AI-assisted model docs & lineage
in-progressAuto-generates documentation for BigQuery models and LookML, suggests data checks, and flags downstream impact when a schema changes — closing the doc gap that always lags model velocity.
KPI anomaly explainer
in-progressWhen a headline KPI moves, an LLM walks the pre-computed segment breakdowns and drafts a plain-English explanation — turning “something’s weird” into a starting hypothesis.
Eval harness for conversational analytics
conceptMethodology write-up on how to evaluate a “talk to your data” product — gold-question sets, semantic similarity scoring, and the failure modes that look correct.
Selected delivery · Verizon
Governance, forecasting, and the team behind it.
The "Big 3" reporting governance
shippedArchitected the three reporting pillars used as the daily governance system for the Indirect channel — covering executive, partner, and aligned-performance views — for 5,000+ partner locations.
XGBoost forecasting for FIOS & FWA
shippedReplaced manual forecasting workflows for FIOS and FWA with XGBoost-based models — significantly improving accuracy while reclaiming the team’s calendar.
Analytics Center of Excellence
shippedEstablished a cross-enterprise Analytics CoE to train and benchmark analytic skillsets — turning ad-hoc tribal knowledge into a repeatable practice for new and existing analysts.