SHT. 01 OF 04 SCALE 1:1
DWG. CC-2026-001 PERSONAL PROFILE
mostly human

Casey
Capshaw

// husband · father · crafter · thinker · conflicted technophile

We're in the midst of the biggest disruption for humanity since the industrial revolution. We're back to the drawing board in nearly all aspects of society. Contact me for projects rethinking small business organization, enhancing human performance, and the way we will work together and find meaning in an AI-centric world.

ai workflows + org design
human dynamics
pragmatic vision
first principles thinking
colorado
SECTION A
caseycapshaw.com REV. A — 2026
00 — AI DEVELOPMENTS
00 / LIVE FEED — REFRESHED DAILY

AI Developments

TODAY IN AI 2026-04-18

Today's AI developments focused on advancing reinforcement learning methodologies. In "Reinforcement Learning via Value Gradient Flow," researchers explored behavior-regularized reinforcement learning approaches that incorporate reference distribution constraints. This technique prevents policy divergence from established baseline policies or datasets, addressing a key challenge in RL where agents can develop unstable or unpredictable behaviors during training. The value gradient flow framework offers a mathematical foundation for maintaining policy stability while still allowing for meaningful learning and improvement. This work contributes to making reinforcement learning systems more reliable and controllable, particularly important for real-world applications where maintaining some level of predictability is crucial. The research addresses the fundamental trade-off between exploration and maintaining reasonable behavior bounds, which has been a persistent challenge in deploying RL systems in practical settings where complete deviation from known-good policies could be problematic or dangerous.

THE WEEK'S ARC 2026-04-11 — 2026-04-18

The past week showcased significant advancements across major AI companies, with particular emphasis on multimodal capabilities and enterprise applications. Google DeepMind dominated with extensive Gemini 3.1 releases, introducing Flash TTS for expressive speech synthesis, Flash Live for natural voice interactions, and Deep Think for scientific reasoning. The company also unveiled Gemma 4 open models, Lyria 3 Pro for extended music generation, and Project Genie for interactive world creation, alongside D4RT for 4D reconstruction and improved Veo 3.1 video generation. OpenAI countered with major product updates including enhanced Codex apps with computer use capabilities, the specialized GPT-Rosalind for life sciences, and GPT-5.4-Cyber for cybersecurity through their Trusted Access program. Anthropic joined the competition by announcing multiple Claude model releases including Opus 4.7, Opus 4.5, and Sonnet 4.6, while establishing new partnerships with Mozilla for Firefox security and expanding their Asia-Pacific presence with a Sydney office. The research community contributed significantly with developments in reinforcement learning through value gradient flow methods, automated webpage design with MM-WebAgent, medical AI through RadAgent for CT scan interpretation, and various studies on LLM spatial reasoning and emotion recognition capabilities.

see sources
  • Google DeepMind unveiled multiple Gemini 3.1 models including Flash TTS for expressive speech, Flash Live for natural voice interactions, and Deep Think for scientific reasoning → Gemini 3.1 Flash TTS: the next generation of expressive AI speech
  • Google DeepMind released Gemma 4 open models, Lyria 3 Pro for extended music generation, and experimental Project Genie for interactive world creation → Gemma 4: Byte for byte, the most capable open models
  • OpenAI released enhanced Codex apps with computer use and image generation capabilities → Codex for (almost) everything
  • OpenAI launched the specialized GPT-Rosalind model for life sciences research → Introducing GPT-Rosalind for life sciences research
  • OpenAI introduced GPT-5.4-Cyber for cybersecurity defense through expanded Trusted Access program → Trusted access for the next era of cyber defense
  • Anthropic announced multiple Claude model releases including Opus 4.7, Opus 4.5, and Sonnet 4.6 → Introducing Claude Opus 4.7
  • Researchers developed value gradient flow methods for behavior-regularized reinforcement learning → Reinforcement Learning via Value Gradient Flow
  • MM-WebAgent was developed for automated webpage design → MM-WebAgent: A Hierarchical Multimodal Web Agent for Webpage Generation
  • RadAgent was created as an AI tool for medical CT scan interpretation → RadAgent: A tool-using AI agent for stepwise interpretation of chest computed tomography
Pipeline last ran: 7h ago
01 — WRITING
01 / SUBSTACK — 8-TRACK TO AI

Writing

MAR 18, 2026
The Immortal Employee
What if the smartest person on your team never actually left? On AI, institutional knowledge, and what it means to retain expertise.
MAR 2026
MAR 11, 2026
The Case for Optimism
Why hope is combat, not surrender. A GenX take on staying constructive in the middle of a technological transition most people still don't believe is real.
MAR 2026
FEB 24, 2026
The Napster Moment for AI
The revolution went open source. The gatekeepers don't stand a chance — and we've seen this movie before.
FEB 2026
FEB 22, 2026
The Stream Deck Nobody Asked For
How a bedtime phone session turned into a GitHub repo by morning. On building with AI, late-night itch-scratching, and shipping before breakfast.
FEB 2026
All Articles ↗
02 — BUILDING
02 / OPEN SOURCE

Building

★ 3
TypeScript · AI Agents
OperatorOne
AI-agent-centric operations platform for solopreneurs and small teams. Brings agentic workflows to everyday business operations.
Public View repo →
★ 1
Python · Developer Tools
streamdeck-claude-code
Stream Deck icon pack and config generator for controlling Claude Code via Ghostty. Hardware shortcut layer for AI-assisted development.
Public View repo →
03 — WORKING
03 / EXPERIENCE

Working

2026 — PRESENT
Strata Identity
Director of Engineering Operations
AI agentic workflows, context engineering for business applications, AI prototyping, and organizational design. Leading the intersection of people systems and AI-native operations.
AI Workflows Context Engineering AI Prototyping Org Design SaaS Ops
current role
2024 — 2026
Strata Identity
Director of Operations & People
Strategic advisor to executive team. Led ADR framework for employee engagement, retention, and leadership development. Implemented scalable systems for conflict resolution and culture-building.
People Strategy Leadership Dev Retention
2021 — 2024
Strata Identity
Director of Operations / Director of Customer Success
Orchestrated global operational systems and SaaS tool implementations (HubSpot, Salesforce). Established and scaled Customer Success for enterprise clients, improving engagement, retention, and NPS.
Operations Customer Success HubSpot Salesforce
2019 — 2020
Accenture
Tech Advisory Consulting Manager
Partnered with senior leadership of a major telecom on organizational design, change management, and Agile implementations. Led Salesforce migration impact assessment.
Consulting Org Design Change Mgmt Agile
04 — CONTACT
04 / CONTACT

Get in touch.

Substack ↗ LinkedIn ↗ GitHub ↗ X ↗