# DEIXEI — Marcio Parente > The site of Marcio Parente, author and engineering leader. It publishes the AI Delivery Factory reference (how to design, build, govern, deliver and operate AI-enabled software in regulated enterprises), the DEIXEI Research Series working papers, books on AI operating models for regulated finance, and a blog on engineering leadership and enterprise software delivery. Every link below to a `.md` file is a Markdown copy of the page (the HTML page is the same URL with `/` in place of `.md`). Each Markdown copy starts with a header giving the page's canonical URL, type and dates. Links inside the Markdown are absolute. Full text by section, one file each: - [AI Delivery Factory, full text](https://www.deixei.com/ai-delivery-factory/llms-full.txt): every reference page (51). Large. - [Research Series, full text](https://www.deixei.com/research/llms-full.txt): every working paper (10). Large. - [Blog, full text](https://www.deixei.com/blog/llms-full.txt): every post (31). - [Core summary](https://www.deixei.com/llms-full.txt): this index with every summary and paper abstract in full, plus the reference's concept pages in full (about 50k tokens). Where a reference page labels its claims, *source claim* means a book or paper states it, *author model* means a framework the sources propose (not an industry standard or a supervisory expectation) and *recommendation* is the page's own synthesis. Named firms such as "Bank D" are composite examples, not real institutions. ## About - [Home](https://www.deixei.com/): Marcio Parente: author, engineering leader and advisor. - [About Marcio Parente](https://www.deixei.com/about/): Engineering leadership, cloud architecture, enterprise software delivery and technology consulting. - [Advisory and consulting](https://www.deixei.com/connect/): Board advisory, executive consulting, interim leadership and workshops on technology strategy and enterprise software delivery. ## Books - [Books](https://www.deixei.com/books/): Everything Is Green, The Regulated AI Operating Model five-volume series, and Enterprise Software Delivery: A Roadmap for the Future. - [The Regulated AI Operating Model](https://www.deixei.com/regulated-ai-operating-model/): Five volumes, by Marcio Parente and Roberto Barbosa, for CIOs, CTOs, CROs and boards in regulated finance who need to run AI agents safely and accountably at scale. Vol I The Adoption Gap, Vol II The Governance Spine, Vol III The Platform Foundation, Vol IV The Agent Operating Model, Vol V Building and Shipping. - [Everything Is Green](https://www.deixei.com/books/): Why AI Is Taking Software Delivery Out of Your Control — and How to Take It Back. By Marcio Parente and Roberto Barbosa. - [Enterprise Software Delivery: A Roadmap for the Future](https://www.deixei.com/book/): Eight years of research into DevOps industrialisation. ## AI Delivery Factory - [AI Delivery Factory reference](https://www.deixei.com/ai-delivery-factory/): A task-oriented technical reference for AI-enabled software in regulated enterprises. Only pages that have completed publication review are listed. - [Reference index (JSON)](https://www.deixei.com/ai-delivery-factory/index.json): pages, sections, provenance, related pages, research papers and tool specifications. ### Understand The factory model and the parts that make it work. - [The AI Delivery Factory: Operating Model](https://www.deixei.com/ai-delivery-factory/architecture/factory-model.md): The factory's six connected parts, the control gap it closes, the handover curve that governs what agents may take over, and how to start one without rebuilding your tooling. - [The Anatomy of an AI Application](https://www.deixei.com/ai-delivery-factory/architecture/application-anatomy.md): A six-tier view of what one AI product is made of, crossed by a trust and operations plane that attaches controls to every tier; what the factory already supplies per tier, how the anatomy maps to the six-layer product… - [The Enterprise Agent Control Plane](https://www.deixei.com/ai-delivery-factory/architecture/enterprise-control-plane.md): The firm-owned services that make every agent action identifiable, decidable, approvable and reconstructable: the ownership boundary, six planes and their contracts, the Policy Decision API with its latency budget, the… ### Design Choose a product shape, agency class, and recipe. - [Run a Discovery-Paced Portfolio](https://www.deixei.com/ai-delivery-factory/design/discovery-portfolio.md): How to decide which problems become products when construction is cheap: the selection rule and its evidence, the nine-field bet card, the kill clause, a retirement recipe run before promotion, and the promotion… - [Build a Living Product Bill of Material and Recipe](https://www.deixei.com/ai-delivery-factory/design/bom-and-recipes.md): Describe what an AI product contains and how the firm builds it in one versioned specification shared by pipelines, agents and assurance teams, including the classification line, readiness grades and reuse generations. - [Choose the Shape of an AI Product](https://www.deixei.com/ai-delivery-factory/design/choose-ai-product-shape.md): A least-agency decision sequence for choosing a product design: agency classes and their trade-offs, the eight questions, evidence for every step up, the consequence matrix that caps action scope, the cut-set review… - [Classify Use Cases into Risk Tiers](https://www.deixei.com/ai-delivery-factory/design/risk-tiers.md): The T0–T4 use-case taxonomy as a working decision reference: the prohibited-use gate, four intake questions, data and action boundaries per tier, the twelve-area control matrix, retention and payload rules, runtime… ### Build the platform Landing zone, identity, data, gateway, connectors and the agent workstation. - [Assess Agent Readiness Across the Estate](https://www.deixei.com/ai-delivery-factory/build/agent-readiness.md): Check the estate's control foundations, grade the stations a change must cross on evidence, route work by the weakest station, fund readiness worst-first with a stated split, and make bought products machine-operable… - [Design the Cloud Production Line and Hosting Boundaries](https://www.deixei.com/ai-delivery-factory/build/cloud-and-hosting.md): Classify, segment, constrain and observe every workspace at the plane where actions execute, choose hosting routes per request, and keep vendor maturity, contracts and exit equivalence dated and explicit. - [Identity, Secrets and Configuration for Agent Workloads](https://www.deixei.com/ai-delivery-factory/build/identity-and-secrets.md): Give every actor one owned identity, size its authority for the worst case, separate runtime from deployment, federate instead of holding keys, refuse impersonation and silent fallback, govern configuration by class… - [Prepare a Governed Landing Zone for AI Delivery](https://www.deixei.com/ai-delivery-factory/build/landing-zone.md): Design the management-group hierarchy, enforce guardrails in deny mode, give humans and agents the same paved workspace, integrate the eight agent-platform components, close the perimeter, and prove the zone is done… - [Build the Standard Application Foundation](https://www.deixei.com/ai-delivery-factory/build/application-foundation.md): Run the shared identity, telemetry, deployment, data and AI capabilities as a funded product that humans and agents both start from, with a contract, evidence-computed trust, a cheap promotion boundary and reuse… - [Coding-Agent Workstation Baseline (Claude Code)](https://www.deixei.com/ai-delivery-factory/build/coding-agent-workstation.md): Put Claude Code on developer workstations and build agents as privileged automation: one device-delivered managed policy per surface, lock flags that stop rules merging upward, command hooks for enforcement and HTTP… - [Govern APIs, Data Products and Retrieval Agents Can Trust](https://www.deixei.com/ai-delivery-factory/build/data-api-integration.md): Turn interfaces and data into owned, versioned capabilities that applications and agents consume under identity, entitlement and evidence controls, and give agents access to meaning, limits and authority, not only bytes. - [Data Classes, Processing Lanes and Tokenisation](https://www.deixei.com/ai-delivery-factory/build/data-classes-and-lanes.md): Classify every model request by its data, route it only to a processing lane the class permits, keep derived artefacts inside their source class, and treat retention, roles and tokenisation as recorded, testable… - [Govern MCP Connectors Through a Broker](https://www.deixei.com/ai-delivery-factory/build/mcp-and-connector-broker.md): Admit, tier, pin and enforce every MCP server through one registry and two enforcement points, keep tokens audience-bound, govern A2A delegation through its own registry and gateway, and expose client-facing agent… - [Choose and Contract Claude Deployment Routes](https://www.deixei.com/ai-delivery-factory/build/deployment-routes.md): Decide, per use-case class, which route carries Claude into production (Anthropic API, Claude Platform on AWS, Bedrock, Google Cloud, Microsoft Foundry, plus the Microsoft 365 Copilot surface); record who operates… - [Design and Operate the Model Gateway](https://www.deixei.com/ai-delivery-factory/build/model-gateway.md): Build the model gateway as the governed edge for every model call: per-client identity, per-request highest-class routing, tokenisation, pinned models, a conformant wire contract, a correlated evidence record, a tested… ### Deliver changes The agent-speed SDLC, release engineering and change governance. - [The Agent-Speed SDLC](https://www.deixei.com/ai-delivery-factory/delivery/agent-speed-sdlc.md): A committed artefact chain, governed context, trusted tests and eight SDLC controls that let coding agents work quickly while intent, review, identity and evidence stay connected. - [Change Governance for Agent-Enabled Delivery](https://www.deixei.com/ai-delivery-factory/delivery/change-governance.md): Route agent-speed change by risk without losing the approver: control values that always reach a person, a routing gate that reads the right signals, autonomy earned per change family, a Virtual CAB that prepares a… - [SDLC Component and Capability Catalogue](https://www.deixei.com/ai-delivery-factory/delivery/sdlc-catalogue.md): The reusable starter-kit components and platform-shared capabilities every new application should inherit, grouped into four SDLC regimes, with the conventions to fix once, provider equivalents, owners, the mapping to… - [Release Engineering for Agent-Written Change](https://www.deixei.com/ai-delivery-factory/delivery/release-engineering.md): Move agent-speed change into production through five evidence-emitting stations: one change type per change, a reversal decided at authoring, a valid Deployer Manifest, a gateway that cannot be bypassed, hard health… ### Govern and prove Controls, regulation, evidence, accountability, models and third parties. - [Build a Unified Control Map](https://www.deixei.com/ai-delivery-factory/govern/control-map.md): Join the firm's AI and technology controls in one traceable operating view: rows are operating concerns, columns are frameworks read at their own altitude, every cell points to live evidence, and existing ITIL, COBIT… - [Route Regulatory Questions to Obligations and Evidence](https://www.deixei.com/ai-delivery-factory/govern/regulatory-map.md): A dated, article-level reference that routes AI design questions to the EU AI Act, DORA and its RTS, GDPR and sectoral rules, and turns each obligation into a control, an owner and evidence: change records under RTS… - [Regulatory Calendar for AI in EU Financial Services](https://www.deixei.com/ai-delivery-factory/govern/regulatory-calendar.md): The stop-the-press calendar rebuilt as a dated register: AI Act stages after the AI Omnibus, DORA and its RTS/ITS, CTPP designations, ESA and supervisor guidance, national laws such as Italy's Law 132/2025 with… - [The Audit Spine and the Evidence Record](https://www.deixei.com/ai-delivery-factory/govern/audit-evidence.md): How to capture, join, protect, retain and prove per-action evidence: the AES v0.1.0 record contract, three capture points and one correlation ID, tiered immutable storage, retention and legal hold, replay rehearsal… - [Model Governance: Usage Registry and Upgrade Lifecycle](https://www.deixei.com/ai-delivery-factory/govern/model-governance.md): Govern models at the grain regulators ask about: a catalogue for what exists, an 18-field usage registry for what is used, risk classified per usage, a registry the policy decision point reads by usage_id, pinned model… - [Close the Policy, Audit and Actuals Loop](https://www.deixei.com/ai-delivery-factory/govern/policy-audit-actuals.md): Run governance as a standing comparison of declared policy against emitted records, route only the exceptions to the second line, and govern the policy packs that drive the comparison. - [Accountability and Ownership](https://www.deixei.com/ai-delivery-factory/govern/accountability.md): Hold provenance and accountability apart on purpose: four named accountabilities per product, a human owner for every agent change, seven named seats, a signed Build/Run RACI and a first-line/second-line boundary that… - [Vendor Concentration and Exit](https://www.deixei.com/ai-delivery-factory/govern/vendor-concentration-and-exit.md): Measure concentration on three axes against 70% and 85% discussion triggers, keep a tested successor for every material model usage, record eight contract-cleavage points per route, prove evidence portability with a… - [The Authority Process](https://www.deixei.com/ai-delivery-factory/govern/authority-process.md): Generative output is never a record. Label drafts, cite their sources, and turn a draft into an authoritative record only through one owned, eight-step process per record class: validation, business rules, data… ### Operate and assure Run the factory and its agents, respond to incidents, assure and measure. - [Operate the Agent Lifecycle](https://www.deixei.com/ai-delivery-factory/operate/agent-lifecycle.md): Inventory, register, sign, zone, review, suspend, recover and retire agents through a six-stage lifecycle with named human ownership, zone-specific authority and evidence at every transition. - [Respond to Agent Incidents](https://www.deixei.com/ai-delivery-factory/operate/incident-response.md): Sort an agent failure into one of four incident shapes, suspend it in three verified layers, keep humans on every classification and disclosure decision, route one playbook to every authority that needs a cover sheet… - [Run the AI Delivery Factory](https://www.deixei.com/ai-delivery-factory/operate/running-the-factory.md): Operate demand, capacity, the two worker classes, exceptions and evidence as one service: read two leading numbers, keep a handover backlog, gate construction on three conditions, run queues and alerts at agent volume… - [Measure the Factory](https://www.deixei.com/ai-delivery-factory/operate/factory-measurement.md): Read usage first, then the seven-dimension scorecard; emit a served-request event that separates human from agent callers; label every reported number by evidence tier; watch the two leading numbers of a control gap… - [Agent-Driven Departments](https://www.deixei.com/ai-delivery-factory/operate/agent-driven-departments.md): What an Agent-Driven Department is and how to test for one, the six-system stack it runs and the order to build it, the measured agent share, and the consolidation boundary: what moves into the department, what stays… - [Evaluate, Assure and Test Agent Resilience](https://www.deixei.com/ai-delivery-factory/operate/evaluation-and-resilience.md): Define a versioned evaluation contract, test the trajectory and the execution boundaries, run three assurance layers into one quarterly pack, map the OWASP Agentic Top 10 to controls that demonstrably fire, bring… ### Adopt and fund Plan the roadmap, readiness, sourcing, budget and charter. - [Readiness Diagnostic](https://www.deixei.com/ai-delivery-factory/adopt/readiness-diagnostic.md): Four self-assessment instruments at four altitudes: the six-question CIO self-diagnostic (can the firm govern agents at all?), the ten-pillar control-plane scorecard (0–60), station readiness (lowest of four, 0–3), and… - [The Implementation Roadmap](https://www.deixei.com/ai-delivery-factory/adopt/implementation-roadmap.md): Sequence the factory as a dependency graph, not a calendar: the closure order for the governance gap, a written portfolio threshold, six overlapping phases over eighteen months with entry conditions, exit evidence… - [The Factory Charter](https://www.deixei.com/ai-delivery-factory/adopt/factory-charter.md): The one ratified document the factory's organisation, business case and operating model hang off: mission, mandate and decision rights, operating principles and ordered values, each value with an owner, policy, metric… - [Build, Configure or Consume](https://www.deixei.com/ai-delivery-factory/adopt/build-configure-consume.md): The layered sourcing decision: a signed portfolio threshold (8 products, 24 months, 3 shared patterns including shared data) decides whether building is available at all; Research Paper 7's four gates decide each… - [Control Budget](https://www.deixei.com/ai-delivery-factory/adopt/control-budget.md): An author proposal for a CRO-signed limit on how fast change may arrive relative to the firm's capacity to understand it: the CIO runs the line, the CRO signs the band, only changes with a full record count, two weeks… ### Patterns Reusable design choices with explicit trade-offs. - [Workflow Pattern](https://www.deixei.com/ai-delivery-factory/patterns/workflow.md): Use a workflow when code can determine the next step, even when some steps call a model: the eight workflow patterns with their forces and factory notes, why class 2 is the default, how code-orchestrated multi-agent… - [Bounded-Agent Pattern](https://www.deixei.com/ai-delivery-factory/patterns/bounded-agent.md): A class 3 agent lets a model choose the next step, but only inside a registered, validated and enforced envelope of Mandate, Means, Memory and Monitoring. - [The Ten AI Product Recipe Families](https://www.deixei.com/ai-delivery-factory/patterns/recipe-catalogue.md): The ten recipe families and their design cards (flow, action scope, data classes, controls, evidence, evaluation, anti-patterns), how to fit a request and choose between neighbours, the pattern catalogue beneath them… ### Worked examples See the decisions applied to a concrete case. - [Worked Example: Research Assistant](https://www.deixei.com/ai-delivery-factory/examples/research-assistant.md): A composite Asset Manager X request is split into a grounded assistant and a deterministic routing workflow instead of a multi-agent research crew, then written down twice: as the ten-field design record and as a… - [Worked Example: A Migration Through the Factory](https://www.deixei.com/ai-delivery-factory/examples/migration-case.md): A composite data-centre exit walked as a reusable decision trail: purpose sheet, recipe boundary, inventory as a discovery control, the control loop, transform review with adversarial cases and business reconciliation… - [Worked Example: An Inspection Answered From the Record](https://www.deixei.com/ai-delivery-factory/examples/inspection-case.md): Composite supervisory walkthrough: for each question a supervisor asks about automated approval and agent actions, the evidence that answers it, the query that produces it, who runs it, and the questions the record… ### Reference Glossary, schemas and tools, primary sources and the source index. - [Glossary and Vocabulary Crosswalk](https://www.deixei.com/ai-delivery-factory/reference/glossary.md): The controlled vocabulary of the AI Delivery Factory reference, with a crosswalk for the places where the books, working papers and specifications use different words for the same thing. - [Schemas, Conformance Packs and the ops Toolchain](https://www.deixei.com/ai-delivery-factory/reference/schemas-and-tools.md): What the versioned Supporting Tools schemas hosted under /aes/v0.1.0/ require, how to validate against them, where the source specification contradicts itself, and what the planned ops CLI, Claude plugins and service… - [Primary Sources and Open Checks](https://www.deixei.com/ai-delivery-factory/reference/primary-sources.md): The official texts this reference relies on (EU legal instruments with ELI links, DORA Level 2 acts, Commission and ESA publications, national law, standards, vendor documentation and the specifications hosted on this… ## Research Series - [DEIXEI Research Series](https://www.deixei.com/research/): Working papers. Each paper also has a short alias, /research/wp-NN. - [AI Reference Architectures and Design Patterns: A Taxonomy for a Reuse-Driven AI Factory (Working Paper 01)](https://www.deixei.com/research/taxonomy-for-a-reuse-driven-ai-factory.md): No single published AI reference architecture is sufficient for a regulated financial institution that wants to deliver AI applications repeatedly instead of one project at a time. Hyperscalers describe lifecycle and… - [Ten Solution Archetypes for Regulated Finance (Working Paper 02)](https://www.deixei.com/research/ten-solution-archetypes-for-regulated-finance.md): This paper proposes ten solution archetypes as the starting points for AI applications in a regulated financial institution: a bank, an insurer or an asset manager in the EU, the UK or Switzerland. An archetype is a… - [Securing Claude Code in the Enterprise (Working Paper 03)](https://www.deixei.com/research/securing-claude-code-in-the-enterprise.md): Claude Code is an agentic coding tool that reads files, runs shell commands, edits code and calls external tools for a developer or a pipeline. That makes it a different security object from a chat assistant. This… - [Technology Management and Agent Readiness: Why the Production Line Runs at the Speed of Its Least-Ready Shared Station (Working Paper 04)](https://www.deixei.com/research/technology-management-and-agent-readiness.md): Technology management — the technology radar and lifecycle, version discipline, platform standardisation, deprecation and golden paths — has long been funded as hygiene. This paper argues that it now decides how much… - [Mission and Values of the AI Factory: What Changes When the Values Document Is Executed (Working Paper 05)](https://www.deixei.com/research/mission-and-values-of-the-ai-factory.md): Mission and values statements were written for people, who interpret them, discount them and, when the statement contradicts the incentive, follow the incentive. In a delivery organisation where AI agents do much of… - [The AI Factory versus the Software Factory: What Nearly Sixty Years of Factory Ambition Teach About Product Development at Agent Speed (Working Paper 06)](https://www.deixei.com/research/ai-factory-versus-software-factory.md): The software factory is an idea from 1968 that has been attempted in every decade since: at System Development Corporation, in the Japanese factories Michael Cusumano documented, in CASE and model-driven development… - [Build, Configure or Consume: Sourcing Decisions When Agents Make Code Cheap (Working Paper 07)](https://www.deixei.com/research/build-configure-or-consume.md): AI coding agents have made code cheap to produce, and a common inference follows: build everything as code. This paper tests that inference against the evidence published in 2025 and 2026 and finds that it does not… - [Reusability and the Compounding Factory (Working Paper 08)](https://www.deixei.com/research/reusability-and-the-compounding-factory.md): The business case for an AI factory in a regulated financial institution rests less on agents writing code faster than on a few shared assets whose value grows with every product team that consumes them. I argue that a… - [Data Classes, Processing Lanes and Controller/Processor Roles for AI (Working Paper 09)](https://www.deixei.com/research/data-classes-processing-lanes-and-roles.md): A language model has to see data to work on it, and it often sees that data on somebody else's infrastructure. For a regulated financial institution this turns two governance questions into design-time questions: how… - [MCP and A2A: Interoperability Protocols for the Enterprise AI Factory (Working Paper 10)](https://www.deixei.com/research/mcp-and-a2a-interoperability.md): Two open protocols now carry most of the interoperability work in enterprise agentic AI. The Model Context Protocol (MCP) standardises how an AI application reaches tools and data. The Agent2Agent Protocol (A2A)… ## Machine-readable resources - [Audit Evidence Schema v0.1.0](https://www.deixei.com/aes/v0.1.0/audit_evidence_schema.json): JSON Schema for agent action evidence records. - [CAB change record v0.1.0](https://www.deixei.com/aes/v0.1.0/cab_change.json) and [CAB decision record v0.1.0](https://www.deixei.com/aes/v0.1.0/cab_decision.json): JSON Schemas for change dossiers and decisions. - [Deployer Manifest v0.1.0](https://www.deixei.com/aes/v0.1.0/deployer_manifest.json): JSON Schema for gated artefact-to-target deployments. - [Worked examples](https://www.deixei.com/aes/v0.1.0/examples/cs_v2.4.0_manifest.json): example records under /aes/v0.1.0/examples/ (credit_scoring_action.intoto.json, credit_scoring_action.predicate.json, cs_v2.4.0_change.json, cs_v2.4.0_decision.json, cs_v2.4.0_manifest.json). - [Supporting Tools source](https://github.com/deixei/Building-Organizations-for-the-Age-of-AI/tree/master/05_supporting_tools): versioned specifications and implementation documentation. - [Sitemap](https://www.deixei.com/sitemap-index.xml): every HTML page. ## Blog - [Blog](https://www.deixei.com/blog/): Articles and updates. - [Everything Was Green](https://www.deixei.com/blog/2026-11-14-everything-was-green.md): A four-line change, fourteen green tests and an automated approval: how a firm in which every control said yes still let a reconciliation control fail for nine weeks, and what we would check in our own routing before… - [Evidence a Supervisor Can Query](https://www.deixei.com/blog/2026-11-14-evidence-a-supervisor-can-query.md): When the supervisor's letter arrives, the time goes on translation between systems, and translation degrades evidence. Why the one investment that pays twice is a record written by the act it describes, at the moment… - [Provenance Is Not Accountability](https://www.deixei.com/blog/2026-11-14-provenance-is-not-accountability.md): A perfect trail of what your agents did is not the same as a person who will answer for it. Why accountability evaporates when authorship moves to an agent, and the three records a regulated firm has to keep on purpose. - [Where to Start: The First Ninety Days](https://www.deixei.com/blog/2026-11-14-where-to-start-the-first-ninety-days.md): Most AI programmes start with the platform because platform work looks like progress. The first ninety days should start somewhere less visible: with an afternoon, a pencil, two numbers and the order in which things… - [New Series: The Regulated AI Operating Model](https://www.deixei.com/blog/2026-08-02-the-regulated-ai-operating-model.md): My new five-volume series with Roberto Barbosa is out: an operating model for running AI agents safely and accountably in regulated finance — and still answering the regulator's question. - [Building Teams That Don't Need You: True Leadership](https://www.deixei.com/blog/2026-01-16-building-teams-that-dont-need-you.md): The highest form of leadership is building teams so capable, they don't need you. Here's how to create self-sufficient, high-performing organizations. - [Why Most Digital Transformations Fail (And How to Fix Them)](https://www.deixei.com/blog/2026-01-16-why-digital-transformations-fail.md): After 24 years leading technology transformations, I've seen the same patterns emerge. Here's why most digital transformations fall short—and the strategic framework that works. ## Optional - [Taxonomy: tenant classification](https://www.deixei.com/blog/2023-05-02-taxonomy-tenant-classification.md): I classify Azure workloads as non-production (NP) or production (PR), and use that classification to shape management groups, subscription types and policies. Stages, environments and instances are a separate axis that… - [Taxonomy: metadata schema](https://www.deixei.com/blog/2023-04-06-taxonomy-metadata-schema.md): Enterprise software delivery needs a taxonomy to classify resources and a metadata schema, aligned with it, to describe them. Together they are what makes an everything-as-code approach to automation consistent across… - [Compliance](https://www.deixei.com/blog/2023-04-05-compliance.md): Compliance starts with analysing which regulations and standards apply to a product, then building controls and processes that meet them and keep meeting them. I list the regulations software teams most often meet and… - [Cloud governance](https://www.deixei.com/blog/2023-04-04-cloud-governance.md): Cloud governance is the set of policies, processes and controls that keeps a cloud estate secure, compliant and cost-effective. I treat it like a body of law: published principles, standard procedures, and a compliance… - [Creating reference architectures](https://www.deixei.com/blog/2023-04-03-creating-reference-architectures.md): How I set up the work of creating reference architectures: who is responsible, accountable, consulted and informed, how to derive the architectures from the Azure Well-Architected Framework, and a first basic web… - [Coding dojo and code kata](https://www.deixei.com/blog/2023-03-30-coding-dojo-code-kata.md): Teams get better at hard problems by practising on small ones. I explain what a coding dojo and a code kata are, and why I build regular practice time into how a team delivers software. - [A DevOps engineer's story](https://www.deixei.com/blog/2023-03-26-devops-engineer-story.md): Mary, a DevOps engineer, has to move a tightly coupled on-premises estate to the cloud. A short story about everything as code, automation and avoiding lock-in, and why the hardship of change depends on how well we can… - [Reference architecture, landing zone, blueprint, code templates and scaffolding](https://www.deixei.com/blog/2023-03-28-understanding-concepts-reference-architecture-landing-zone-blueprint.md): Five related concepts and how they build on each other: reference architectures guide, landing zones make an architecture deployable, blueprints implement use cases, and code templates and scaffolding get projects… - [Using Power BI](https://www.deixei.com/blog/2023-03-25-utilization-power-bi.md): Automation needs a window onto the factory floor. Why I use Power BI, alongside Azure and Application Insights, to see the state of automated processes in real time instead of building one-off applications to find out. - [Using Azure DevOps](https://www.deixei.com/blog/2023-03-22-utilization-azure-devops.md): Azure DevOps covers the whole software development lifecycle in one suite: work item tracking, version control, CI/CD, test management and release management. How it differs from GitLab and GitHub in scope, integration… - [Why I prefer Bicep over ARM templates](https://www.deixei.com/blog/2023-03-21-preference-bicep-over-arm.md): ARM templates are powerful but verbose, low-level JSON. Bicep is Microsoft's open-source, declarative language for Azure infrastructure as code: easier to read, write and maintain, and still young in 2023. - [Why I prefer Ansible](https://www.deixei.com/blog/2023-03-18-i-prefer-use-ansible.md): I use Ansible less as an infrastructure-as-code tool and more as a metadata and execution-isolation layer: agentless, idempotent, driven by YAML, and extensible enough to codify recipes for every building block and… - [Using Python](https://www.deixei.com/blog/2023-03-17-use-python.md): Why I use Python as the scripting language around Ansible, Azure CLI and other command-line tools: readable syntax, a large standard library, strong third-party libraries and a big community. - [Azure rather than AWS or GCP](https://www.deixei.com/blog/2023-03-16-azure-rather-than-aws-gcp.md): Why I chose Azure: on price, services and regions the big clouds are close, but Windows, Office 365 and Azure DevOps working together give me the best vertical integration and return for each dollar spent. - [YAML rather than JSON](https://www.deixei.com/blog/2023-03-15-yaml-rather-than-json.md): I declare human inputs, such as bills of materials, in YAML because people can read and review it, while APIs and integrations between processes keep using JSON. - [Technology choices](https://www.deixei.com/blog/2023-03-14-technologys-choices.md): A good technology choice is the one that fits the company's strategy, culture and existing infrastructure, which is not always the one with the most benefits on paper. This post sets out the factors I weigh, for… - [Hyper compute](https://www.deixei.com/blog/2023-03-12-hyper-compute.md): What hyper compute (high-performance computing) is, how it differs from PC architecture and general cloud platforms, what Azure offers for it, and the workloads it suits. - [Configurations, feature flags and settings](https://www.deixei.com/blog/2023-03-10-configurations-feature-flags-settings.md): Configuration changes, feature flags included, alter how a system behaves, so I treat them as changes: versioned, tested, approved and recorded through release management like any other. - [Artificial intelligence in automation](https://www.deixei.com/blog/2023-03-09-artificial-intelligence-automation.md): Where AI could help industrialised software delivery: natural language processing for tickets, logs and documentation today, the theoretical promise of AGI, and AI-assisted code portability. - [Pipeline](https://www.deixei.com/blog/2023-03-07-pipeline.md): A DevOps pipeline is a chain of simple, compatible actions, much like a production line. Why I separate the develop pipeline from the main pipeline, and why mission-critical applications also need an emergency pipeline. - [Inner source](https://www.deixei.com/blog/2023-03-06-inner-source.md): Inner source brings open-source practice inside the company: shared code, open collaboration and decisions on merit. Here are its principles, an example definition of done and code of conduct, and why I think it matters. - [Some old photos](https://www.deixei.com/blog/2022-05-01-some-old-photos.md): A selection from more than two decades of photography: how I started, the cameras and teachers along the way, and why I keep taking photos when I travel. - [Cloud genesis](https://www.deixei.com/blog/2015-09-04-genesis.md): Where I start a cloud strategy on Azure: identity first, then cloud types, automation, taxonomy, and a subscription model that fits the company and its IT organisation. - [Azure DevOps Factory](https://www.deixei.com/blog/2015-09-03-azure-devops-factory.md): You cannot industrialise the cloud without controlled automation. The foundation is architectural metadata, read by an orchestrator that triggers everything else. - [ops v0.1.0 — first reference-implementation release](https://www.deixei.com/releases/cli-v010.md): First versioned release of the ops CLI — the reference implementation of the Agent Operations specification. Bundles milestones M09 through M19, with Sigstore-signed binaries and signed self-update. - [Photography](https://www.deixei.com/photography/): Photography collection by Marcio Parente. - [Privacy](https://www.deixei.com/privacy/): What the site stores about visitors and members, and how to have it deleted.