My profile picture

Senior Solutions Architect — Digital Platforms & AI Systems

I help insurers ship Guidewire transformations end-to-end: ClaimCenter/PolicyCenter delivery, integration architecture, and modern digital experiences with Jutro. I also write and build side projects in ML and production-grade AI agents.
Core insurance platform systems and integration layer diagram.
Digital portals architecture with cloud infrastructure and CI/CD pipeline.
Foundational machine learning visuals: decision boundary, sigmoid curve, contour plot.
Unrolled sequence model diagram showing memory gating and fading gradients.
Reinforcement learning training instability dashboard with reward, entropy, KL spikes.
Trading agent loop from order book features to policy execution and feedback.
Modern web architecture stack from browser and edge to APIs, services, and data.
Operability visuals: CI/CD pipeline, distributed tracing, and transactional outbox pattern.
Agent platform architecture with tool gateway, policy gates, evals in CI, and observability.
Core insurance platform systems and integration layer diagram.
Digital portals architecture with cloud infrastructure and CI/CD pipeline.
Foundational machine learning visuals: decision boundary, sigmoid curve, contour plot.
Unrolled sequence model diagram showing memory gating and fading gradients.
Reinforcement learning training instability dashboard with reward, entropy, KL spikes.
Trading agent loop from order book features to policy execution and feedback.
Modern web architecture stack from browser and edge to APIs, services, and data.
Operability visuals: CI/CD pipeline, distributed tracing, and transactional outbox pattern.
Agent platform architecture with tool gateway, policy gates, evals in CI, and observability.
Core insurance platform systems and integration layer diagram.
Digital portals architecture with cloud infrastructure and CI/CD pipeline.
Foundational machine learning visuals: decision boundary, sigmoid curve, contour plot.
Unrolled sequence model diagram showing memory gating and fading gradients.
Reinforcement learning training instability dashboard with reward, entropy, KL spikes.
Trading agent loop from order book features to policy execution and feedback.
Modern web architecture stack from browser and edge to APIs, services, and data.
Operability visuals: CI/CD pipeline, distributed tracing, and transactional outbox pattern.
Agent platform architecture with tool gateway, policy gates, evals in CI, and observability.
Core insurance platform systems and integration layer diagram.
Digital portals architecture with cloud infrastructure and CI/CD pipeline.
Foundational machine learning visuals: decision boundary, sigmoid curve, contour plot.
Unrolled sequence model diagram showing memory gating and fading gradients.
Reinforcement learning training instability dashboard with reward, entropy, KL spikes.
Trading agent loop from order book features to policy execution and feedback.
Modern web architecture stack from browser and edge to APIs, services, and data.
Operability visuals: CI/CD pipeline, distributed tracing, and transactional outbox pattern.
Agent platform architecture with tool gateway, policy gates, evals in CI, and observability.

About Me

I’m a senior solutions architect and consultant focused on insurance platforms and digital delivery. My day-to-day work spans core-system transformations, integration architecture, cloud delivery patterns, and governance that keeps large programs shippable. Alongside consulting, I run long-form technical writing and side research in ML and agentic systems—especially how to make agents operable in production (evals, tool safety, observability, and compliance-by-design).

Work Experience

2021 - Present

Senior Architect at
Alchemy

2012 - 2021

Senior Consultant at
Guidewire

2010 - 2011

Consultant at
Novabase

Writing & Research

Short essays and long-form series on operable AI agents, evaluation, governance, and production reliability

When Agents Escape the Sandbox

The July 2026 rogue-agent incident turned AI containment from a whiteboard concern into an incident-response problem. This post explains how to design containment, monitoring, kill switches, and forensics for autonomous agents that can operate at cyber speed.

Frontier Model Release Governance

GPT-5.6 made the release pipeline itself the story: restricted access, government review, vetted partners, capability evaluations, and staged rollout. This post explains why shipping a frontier model now looks less like a product launch and more like a national-security workflow.

Search Becomes an Agent Runtime

Google’s Gemini Spark and AI Mode point to a bigger shift: Search is no longer just retrieval plus ranking. It is becoming a runtime for synthesis, generated UI, monitoring agents, and action. This post explains the architecture of actionable retrieval — and the reliability contracts needed when search starts doing things.
Axel brought clarity to a complex platform and integration landscape. He translated ambiguity into an execution plan with clear contracts, ownership boundaries, and delivery guardrails. The team shipped with fewer surprises and a calmer release cadence.
Program Lead (anonymized)

Program Lead (anonymized)

Program Lead, Insurance Platform Transformation

Axel consistently reduced delivery friction across teams. He made architectural decisions actionable, aligned stakeholders quickly, and improved reliability through pragmatic release and operational discipline.
Delivery Manager (anonymized)

Delivery Manager (anonymized)

Delivery Manager, Enterprise Program

Axel operates as a force multiplier. He can go from system-level architecture to hands-on delivery support, and he leaves teams with reusable patterns—not just diagrams.
Engineering Lead (anonymized)

Engineering Lead (anonymized)

Engineering Lead, Digital & Integrations

Frequently Asked Questions

Answers to common questions about my process and services.
Axel Domingues - 2026