Shipping
Jul 2A11yReadyThe self-improving loop now reaches the pipeline's prompts, not just its code. Many…Jun 30A11yReadyThe system now diagnoses and fixes its own recurring problems. The Pipeline Advisor was…Jun 29Testing PlatformSelf-serve admin-perms diagnostic landed: 4 role codes unblock 32 specs, verified…Jun 29Testing PlatformThe platform automatically surfaced 5 real legacy bugs in UCR QAT — including 3 that…Jun 29Testing PlatformHonest Monday baseline: 66.2% pass excl. skips on the cleaned 210-spec / 324-test suite…Jun 29Testing PlatformThe strict PageObject allowlist now actually constrains the live generator. The v3.2…Jun 28A11yReadyTwo reliability improvements. (1) Image sizing: images inside converted documents —…Jun 28File IntelligenceRetired the Xingwu PHP-audit cross-check — it was the initial one-time orphan-detection…Jun 28Testing PlatformEight of nine holistic test categories now populated on prod (was six Friday). Sunday…Jun 27A11yReadyLibrary redesigned as a compact, sortable list (was large cards) — far more documents…Jul 2A11yReadyThe self-improving loop now reaches the pipeline's prompts, not just its code. Many…Jun 30A11yReadyThe system now diagnoses and fixes its own recurring problems. The Pipeline Advisor was…Jun 29Testing PlatformSelf-serve admin-perms diagnostic landed: 4 role codes unblock 32 specs, verified…Jun 29Testing PlatformThe platform automatically surfaced 5 real legacy bugs in UCR QAT — including 3 that…Jun 29Testing PlatformHonest Monday baseline: 66.2% pass excl. skips on the cleaned 210-spec / 324-test suite…Jun 29Testing PlatformThe strict PageObject allowlist now actually constrains the live generator. The v3.2…Jun 28A11yReadyTwo reliability improvements. (1) Image sizing: images inside converted documents —…Jun 28File IntelligenceRetired the Xingwu PHP-audit cross-check — it was the initial one-time orphan-detection…Jun 28Testing PlatformEight of nine holistic test categories now populated on prod (was six Friday). Sunday…Jun 27A11yReadyLibrary redesigned as a compact, sortable list (was large cards) — far more documents…
Loren Cossette

About

Loren Cossette

Ph.D. (c) · AI Program Architect

I'm an embedded AI program architect. Currently the AI lead for Multnomah County, running 5 production AI initiatives simultaneously across accessibility, site intelligence, legacy modernization, and QA automation — shipping features weekly with a dated public activity feed anyone can verify.

The program reaches from architecture to code: LangGraph pipelines, multi-agent systems on Vertex AI (Claude Sonnet 4.6, Gemini 2.5), RAG over pgvector, deterministic forms parsing, behavioral test capture from retiring SMEs. The same program handles stakeholder enablement: executive briefings, ROI calculators, SME approval queues, and validation against independent external audits.

Outside the County: executive AI training at UT Austin and Johns Hopkins, peer-reviewed AI research, side products (DraftEngine, GTM Teardown, Clearview Politics). PhD candidate. Prosci-certified. SHRM-SCP. 20-year military career culminating at E-9 (top 1%).

~/loren-cossette

$ cat philosophy.md

# I believe the best AI engineers don't just write code —

# they understand the business problem deeply enough

# to know which problems are worth solving with AI,

# and which ones aren't.

# Strategy without code is a slide deck.

# Code without strategy is a science project.

# I do both.

Education

Academic foundation

Ph.D. — Leadership & Program Evaluation

UIW

In Progress

Post Graduate Certificate — AI and Machine Learning

University of Texas at Austin

Post Graduate Certificate — AI/ML, LLM & GenAI

Cornell University

Post Graduate Certificate — Strategic Artificial Intelligence

Houston Christian University

M.S. — Organizational Development

University of the Incarnate Word

M.A. — Psychology

Northcentral University

Certifications

SHRM-SCP
Prosci Change Management
SAFe 5.0 Agilist
EQi-2.0 Practitioner
7φ

Engineering Methodology

My 7-phase workflow from problem discovery to production handoff.

Publications

Peer-reviewed research

2025

Bias and Reliability in AI Safety Assessment: Multi-Facet Rasch Analysis of Human Moderators

AIME

2024

AI Integration in Academia: Investigating the Impact on Faculty Organizational Belongingness

APA

2024

Self-Reflection and Socialization in Early Career Research: A LACOID-Based Text Mining Approach

AERA

2023

Assessing Changes in Adult Online Learners’ Motivation: A Grounded Bibliometric Analysis

Book Chapter

2023

Exploratory Cluster Analysis of Technology Use During COVID-19 and Mental Health

J. Adolescent Health