Experience & Projects

AirOps

Head of Research & Product Intelligence

Dec 2025 — Present

Head of Research studying AI search and RAG systems. I research how LLMs retrieve, select, and generate responses, build ETL pipelines and data models, and work alongside product to turn findings into features.

  • Leads research studies, model evaluations, benchmarks, and agent harness testing to understand how LLMs retrieve, cite, follow instructions, and respond across different user search behaviors.
  • Builds reusable research pipelines for large-scale model testing, including ETL workflows, data models, scoring logic, and evaluation datasets.
  • Turns AI search and RAG findings into product intelligence, prediction models, internal tools, dashboards, and platform features.
Agent Harness Benchmark →Content Quality Evaluation Model
AI Research Lead & Content Engineer Jul 2025 — Dec 2025
  • Led research studies on AI search behavior, retrieval patterns, citation selection, and content quality signals across LLM systems.
  • Designed and validated prompt systems for research workflows, content evaluation, AI visibility analysis, and structured output generation.
  • Turned research findings into org-wide enablement materials, including reports, internal briefs, workflows, and guidance for stakeholders, product, marketing, and customer-facing teams.
Content Engineer May 2025 — Jul 2025
  • Built development and generative workflows using API integrations, structured prompts, JSON outputs, and reusable workflow components.
  • Developed prompt engineering systems for testing LLM generation, controlling output structure, and improving consistency across model responses.
  • Created reusable prompt logic, structured output patterns, and self-executing processes that could be adapted into production components.

Verkada

AI & Search Advisor

Oct 2025 — Feb 2026

Advised on AI search strategy and brand visibility across traditional search and LLM retrieval systems. Focused on site architecture, indexation, crawlability, UX, and technical structure to improve how search engines and RAG systems access, interpret, and retrieve Verkada’s web pages.

Teal

Senior Growth Manager, Organic Search

Jul 2024 — Mar 2025

Led technical growth initiatives across UX, CRO testing, site architecture, and user retention. Partnered with product, design, and engineering to improve acquisition paths, page experience, crawlability, indexation, and conversion performance.

Growth Marketing Manager, Organic Search Mar 2024 — Jul 2024

Guru

Head of SEO

Feb 2023 — Nov 2023

Led organic search strategy across technical SEO, site architecture, web migrations, reverse proxy implementation, and brand visibility. Improved crawlability, indexation, internal linking, entity recognition, and page-level structure across Guru’s web properties.

SEO Lead, Growth Dec 2021 — Feb 2023
01 / 01
Content Quality Data Pipeline 01

15,000-query empirical research pipeline measuring which content quality signals predict ChatGPT citation. Six idempotent pipeline steps collect ChatGPT responses through the real consumer interface, join against Google SERP data for original and fan-out queries, extract 24 content quality signals per page, and run multi-method statistical analysis.

empirical researchstatistical analysislogistic regressionrandom forestproduction research pipeline
view details →
AirOps Research Agent System 02

19-agent, 13-skill multi-agent orchestration system running end-to-end research report production at AirOps. Plugin-style YAML-defined agents with per-agent skill loadouts, three lifecycle maps (full report, micro-report, insight drip), per-agent session memory, and cross-tool MCP integration across Notion, Slack, Asana, AirOps, ClickHouse, and Google Workspace.

multi-agent orchestrationagent architectureMCP integrationsworkflow designresearch operations
view details →
Fan-Out Coverage Study 03

16,877-query observational research pipeline measuring whether ChatGPT cites pages that cover more of its internal fan-out sub-query space, against a deterministic seen-but-not-cited control group. Queue-orchestrated multi-worker pipeline with novel headings-as-index coverage scoring and on-demand section-text embedding.

observational studyembedding methodologyproduction research pipeline
view details →
Agent evaluation benchmark [supplemental] 04

Control-arm execution platform for the AO Next Benchmark scaffolding ablation. Resolves grid inputs into Playbook input variables, renders staged prompts that preserve Playbook stage bodies verbatim, runs them automatically as a single-chat baseline with per-stage instrumentation, and generates harness-vs-control article pair files for downstream human preference labeling.

eval execution platformmodel evaluationprompt cachingscaffolding ablation
view details →
Evaluator Tool Internal 05

Production blind-A/B human evaluation web app for the AO Next Benchmark scaffolding ablation. Same 5+5 criteria rubric and 6-tier scoring scale as the paired LLM-as-judge framework, so human and LLM scores are directly comparable.

human evaluationmodel evaluationblind A/B labeling
view details →
Content Quality Scorer (Tool) 06

Production lead-magnet web tool scoring any article URL across 14 content quality dimensions through a four-stage async pipeline. Multi-method evaluator system (LLM-as-judge, regex / structural parsing, embedding-based scoring) routed across multiple inference providers.

production web toolcontent evaluationmulti-method evaluator
view details →
Claudine Dashboard 07

Local control-plane web UI and background spawner daemon for the AirOps Research Agent System (19 agents). Two-process architecture that polls for queued tasks, spawns agent subprocesses with each agent's declared tool list, and streams output back to a shared store for live in-UI visibility.

control planesubprocess managementoperational tooling
view details →
ChatGPT Internal Query Extractor 08

Browser extension that extracts ChatGPT's internal fan-out search queries from any active conversation. Reuses the user's existing ChatGPT session, walks the conversation message tree to isolate the most recent user prompt and its assistant reply, and exports the fan-out queries to JSON, Markdown, or clipboard.

browser extensionfan-out extractionAI search research tooling
view details →
Sitemap Topic Cluster 09

Python CLI for sitemap-driven topic gap analysis. Lemmatization with POS filtering and a generic-SEO-term blacklist tokenizes URL slugs, groups URLs into topic clusters by token frequency, and surfaces under-covered topics as content gap candidates with example slugs.

NLPtopic clusteringSEO
view details →
RL Agent 10

Reinforcement-learning sandbox with PPO training pipelines for a standard benchmark environment and a custom 1D environment (InterruptGoalLineEnv) where the goal switches at a randomized mid-episode timestep to probe how the agent handles interruptibility.

reinforcement learningPPOcustom environment designinterruptibility
view details →
osaldvha.org 11

Personal portfolio and research site built with Astro and Tailwind. Ethereal brutalist design system with custom canvas scenes, retro terminal UI components, pixel-art assets, ASCII word spirals, film grain overlay, and a content workflow that authors all pages as Markdown in Obsidian.

AstroTailwinddesign systemObsidian workflow
view details →

Let's Connect!

Have a question, an idea you want to discuss, or a rabbit hole worth exploring? Let’s chat about it!