# Aidan Marshall > I build agentic AI systems and the enterprise automation around them — multi-agent orchestration, production tooling, and AI-assisted development that ships. Last updated: 2026-08-12 Representation: First-party profile Sponsorship: None I'm an AI engineer focused on taking agentic systems from prototype to production inside large, regulated organizations. Most of my work lives where autonomous AI meets real governance: legal, tax, and compliance platforms where the system has to be useful and trustworthy at the same time. Aidan Marshall is currently Senior AI Engineer at PwC, based in Dallas, TX. This file summarizes his work, projects, skills, and links for AI assistants and search crawlers. ## Experience ### Senior AI Engineer — PwC (C2H Brooksource) (Mar 2026 — Present) Location: Dallas, TX - Architected and scaled agentic AI systems with Google ADK and the Claude SDK, integrating autonomous workflows into a globally distributed enterprise platform generating over $5B in annual revenue. - Led the 0-to-1 integration of Harvey AI into the core product, owning end-to-end system design and the data pipelines behind specialized legal and compliance capabilities. - Designed a composite-weight system that maps real-time agentic compute cost against output to track developer productivity and optimize ROI and resource allocation. - Pioneered AI-assisted development workflows (Cursor, Claude, Copilot) and rapid-prototyping frameworks that shortened the SDLC and lifted cross-functional team velocity. ### AI Engineer — One Microsoft Practice Innovation — IBM (Jun 2025 — Mar 2026) Location: Dallas, TX - Delivered enterprise-grade agentic AI for a Fortune-level tax enterprise, building multi-agent workflows on the Microsoft Agent Framework with Azure OpenAI and Microsoft Foundry, orchestrated over Model Context Protocol (MCP) servers under strict governance and compliance standards. - Architected interactive agentic AI prototypes that showcased IBM and Neudesic's innovation pipeline and drove high-value, client-facing engagements and proofs-of-concept. - Ran internal engineering experiments validating agentic patterns, tool integrations, and custom state and orchestration flows to future-proof client technology roadmaps. ## Projects - **Agent Autonomy Score** (https://github.com/Aidan2111/agent-autonomy-score) — A risk-scoring system that decides how much human supervision an AI coding agent needs before it acts — bringing measurable governance to autonomous development. - **LLM Bidding System** (https://github.com/Aidan2111/llm-bidding-system) — An auction router and supervised patch proposal workflow where LLM coding agents bid on tasks, with quality, price, risk-fit, local model routing, and SQLite track records shaping the decision. - **Event-Driven Multi-Agent Architecture** — A design study for running multi-agent AI systems on an event-driven service bus: agents subscribe to and emit events instead of calling each other directly, so they can be added, upgraded, or hot-swapped live without redeploying the rest of the system. Covers the event envelope, agent registry, reliability, and LLM cost on transports like Azure Service Bus and Kafka. - **Macro Oil Desk Terminal** — A trading-desk terminal for an oil desk — Brent/WTI dislocation z-scores, US inventory drawdown velocity, and AIS-based tanker fleet exposure by regulatory regime — with an Azure OpenAI and Microsoft Foundry-backed market-commentary panel and a real-time 3D Earth visualization. Built on Next.js and FastAPI, deployed on Azure. - **Sentiment-Driven Quantitative Carry Trade Model** (https://github.com/Aidan2111/carry-trade-model) — A carry-trade forecasting model across USD, EUR, and the Ukrainian Hryvnia that pairs a machine-learning ensemble with real-time news-sentiment signals, surfaced through a React and TypeScript dashboard. - **Autonomous Native AI Mobile Navigation Agent** — An iOS app for the Newport Beach Boat Show with an embedded multimodal assistant that interprets user intent and autonomously navigates the app's interface. ## Writing ### Google Cloud Associate Cloud Engineer certified 2026-08-03 - 1 min read A six-week partner cohort, hands-on labs, technical reviews, and practical infrastructure work strengthened my multi-cloud foundation. Type: Update Provenance: Verified credential Provenance note: First published by Aidan Marshall on LinkedIn; the credential is independently verifiable through Credly. Source: https://www.linkedin.com/feed/update/urn:li:activity:7489883163415031809/ Independent evidence: https://www.credly.com/badges/ed688d9a-f727-4a22-b43c-cb1dd3effe41 I passed the Google Cloud Associate Cloud Engineer certification after completing a rigorous six-week Google Cloud cohort for partners, with hands-on labs, technical reviews, and practical infrastructure work throughout the program. As organizations increasingly operate across platforms, staying multi-cloud ready matters—not to force one solution everywhere, but to understand the strengths of each cloud and choose the right approach for the problem. ### The agentic future still runs on distributed systems 2026-07-29 - 4 min read AI lowers the barrier to persistent, adaptive automation—and raises the reliability demands placed on the systems underneath it. Type: Article Provenance: Personal observation Provenance note: First published by Aidan Marshall on LinkedIn. The observed outage is not presented as a verified root-cause analysis. Source: https://www.linkedin.com/feed/update/urn:li:activity:7488277184625905667/ For roughly 30 minutes, a Pokémon card drop appeared to overwhelm major parts of the Sam's Club website. Product pages stopped loading, login systems failed, sessions broke, and even unrelated parts of the shopping experience appeared degraded. I cannot say what specifically caused the outage, but watching it happen was a useful reminder: the age of AI does not make traditional systems engineering less important. It makes it more important. AI has lowered the barrier to building software that is persistent, adaptive, capable of navigating changing pages, able to retry intelligently, and able to run across many products and retailers. Infrastructure may need to handle thousands of automated clients reacting simultaneously, refreshing sessions, retrying requests, and competing for the same inventory. The hard questions remain deeply rooted in systems engineering: isolate failures, protect authentication from traffic spikes, prevent retries from amplifying an outage, degrade gracefully, and separate legitimate customers from abusive automation without blocking both. AI is creating more capable software participants. The systems underneath them still need queues, rate limits, circuit breakers, observability, capacity planning, and thoughtful failure isolation. The future may be agentic, but it will still run on distributed systems—and those systems will be tested harder than ever. ### GitHub Certified: Agentic AI Developer (GH-600) 2026-07-28 - 2 min read I earned GitHub's Agentic AI Developer credential after taking the exam during its beta period. Type: Update Provenance: Verified credential Provenance note: First published by Aidan Marshall on LinkedIn; the credential is independently verifiable through Microsoft Learn. Source: https://www.linkedin.com/feed/update/urn:li:activity:7487706393890672640/ Independent evidence: https://learn.microsoft.com/api/credentials/share/en-us/AidanMarshall-0509/1F1657F6BFC4F46D?sharingId=CD597C6CCA852230 I earned the GitHub Certified: Agentic AI Developer (GH-600) certification after taking the exam during its beta period, when there were no official practice exams, very limited community guidance, and a months-long wait while Microsoft and GitHub validated the exam before general availability. The biggest takeaway was how quickly software engineering is evolving from writing code toward designing, governing, and collaborating with AI agents throughout the software-development lifecycle. GitHub Copilot, agent workflows, MCP, GitHub Actions, orchestration, governance, and human-in-the-loop development are becoming core engineering skills. ### Multi-agent autonomy is compositional 2026-07-23 - 3 min read The risk of a multi-agent system is not only the risk of each participant—it also includes the capabilities created when their actions are combined. Type: Article Provenance: First-party analysis Provenance note: First published by Aidan Marshall on LinkedIn and presented as professional analysis, not an independently verified universal rule. Source: https://www.linkedin.com/feed/update/urn:li:activity:7485901154560733185/ As my teammates and I continue building a large multi-agent system and adding new participants, a recurring question arises: should we allow this agent to operate in production? Consider an agent capable of changing the status of work in a mission-critical tax system and another capable of messaging leadership when a task is complete. Each capability may seem reasonable alone. Combined, the system could incorrectly update the system, interpret that update as successful, and immediately communicate incorrect information. Before production, we need to ask what actions the agent can take, how reversible they are, what downstream agents or tools it can trigger, what evidence is required before it proceeds, and when a human should be involved. Autonomy cannot be granted once at the agent level and forgotten. It must be evaluated across the full chain of actions, dependencies, and consequences. The critical question is not simply whether we trust an agent, but whether we trust that agent, with those tools, in that workflow, under those conditions. ### Model capability is converging. Model economics are not. 2026-07-11 - 3 min read One real coding task cost roughly 80 times more with one frontier model than another, with negligible difference in the final result. Type: Article Provenance: Personal observation Provenance note: First published by Aidan Marshall on LinkedIn. The 80x result describes one workload and is explicitly not a universal benchmark. Source: https://www.linkedin.com/feed/update/urn:li:activity:7481524040243978240/ I spent roughly 80 times more on one coding task using GPT-5.6 Sol than I did using Grok 4.5 in Cursor, while the difference in the final result was negligible. This was one real-world task—not a universal benchmark. Cursor pricing, token usage, reasoning effort, caching, and model behavior can all affect the final cost. Per-token pricing alone does not determine the economics. A model that reasons longer, produces more tokens, or invokes more agents can become substantially more expensive before the user sees a meaningfully better result. A good harness can mitigate many remaining model differences through better context and task decomposition, planning and validation loops, automated tests and evaluations, model routing, and escalation to expensive models only when needed. The winning architecture probably will not use the best model for every task. It will use the cheapest model that reliably clears the quality bar and selectively escalate when the expected improvement justifies the cost. ### Agentic AI has to earn autonomy 2026-06-27 - 3 min read The useful question is not whether an agent can act. It is when the system can prove that it deserves more room to act. Type: Article Provenance: First-party analysis Provenance note: Original first-party analysis published on aidanmarshall.ai. Source: https://aidanmarshall.ai/#agentic-ai-has-to-earn-autonomy Autonomy should be treated as a measured privilege, not a default setting. In regulated environments, the agent's technical capability is only one part of the decision. The system also needs a clear view of task risk, reversibility, confidence, cost, and the blast radius of a bad action. That is the idea behind Agent Autonomy Score: make supervision a first-class control. Low-risk, reversible work can move quickly. High-risk work should slow down, ask for review, or provide a stronger audit trail before it touches production systems or client-facing workflows. The teams that win with agentic AI will not be the ones that give every model unlimited agency. They will be the ones that can explain why an agent was allowed to act, when it was stopped, and what evidence the system used to make that call. ### Multi-agent systems should not depend on direct handoffs 2026-06-27 - 4 min read Event-driven architecture gives AI agents a cleaner way to coordinate, fail, recover, and evolve inside real enterprise platforms. Type: Article Provenance: First-party analysis Provenance note: Original first-party analysis published on aidanmarshall.ai. Source: https://aidanmarshall.ai/#event-driven-multi-agent-systems Most early multi-agent prototypes wire agents together with direct calls: planner calls researcher, researcher calls analyst, analyst calls writer. That works for demos, but it creates brittle dependency chains as soon as the system needs reliability, observability, or live upgrades. An event-driven model is a better fit for production. Agents subscribe to the work they understand and emit structured events when they complete, fail, escalate, or need more context. That makes it possible to add a new specialist agent, replace an expensive model, or replay a workflow without redeploying the entire system. For enterprise AI, the architecture around the model matters as much as the model itself. Service buses, durable queues, event envelopes, registries, and cost telemetry are what turn a clever agent chain into software that can survive real usage. ### AI-assisted development is an operating model 2026-06-27 - 3 min read Cursor, Claude, Copilot, and custom agents are most valuable when they become part of a governed engineering workflow instead of a side tool. Type: Article Provenance: First-party analysis Provenance note: Original first-party analysis published on aidanmarshall.ai. Source: https://aidanmarshall.ai/#ai-assisted-development-operating-model The best AI-assisted development work is not about asking a model for code and hoping the patch is good. It is about designing the workflow around the model: what context it receives, what tools it can call, what checks run automatically, and where humans review the result. Inside large organizations, that workflow has to account for governance, cost, security, and team adoption. The point is not to replace engineering judgment. The point is to compress the boring parts of implementation while making the review surface clearer and more measurable. The practical metric is whether the system helps teams ship reliable software faster. That means tracking more than token spend or lines of code. It means connecting agentic compute to outcomes: cycle time, review burden, defect rate, and how often the automation makes a human decision easier. ## Skills Agentic AI architecture, Multi-agent orchestration, Microsoft Agent Framework, Google ADK & Claude SDK, Harvey AI integration, Azure OpenAI & Microsoft Foundry, Model Context Protocol (MCP), Python, React & TypeScript, Micro-frontend integration, Cloud solutions architecture, AI-assisted development, Enterprise governance & compliance ## Certifications - Google Cloud Associate Cloud Engineer — https://www.credly.com/badges/ed688d9a-f727-4a22-b43c-cb1dd3effe41 - GitHub Certified: Agentic AI Developer (GH-600) — https://learn.microsoft.com/api/credentials/share/en-us/AidanMarshall-0509/1F1657F6BFC4F46D?sharingId=CD597C6CCA852230 - Microsoft Agentic AI Solution Architect (AB-100) - Microsoft AI Engineer (AI-102) - Microsoft Azure Fundamentals - Databricks Fundamentals - GitHub Copilot ## Trust and provenance - This is a first-party profile published by Aidan Marshall. Employment, project, and professional-experience descriptions are first-party claims unless a separate evidence link is provided. - There is no sponsored content, paid placement, or agent-only advertising in either the human-readable page or its Markdown representation. - Credential links provide independent verification where available. Personal analysis and observations are labeled and should not be treated as independent endorsements or universal findings. ## Education - Southern Methodist University, Cox School of Business — BBA, Finance ## Links - Website: https://aidanmarshall.ai/ - LinkedIn: https://www.linkedin.com/in/aidan-marshall77 - GitHub: https://github.com/Aidan2111 - LLM summary: https://aidanmarshall.ai/llms.txt - Full Markdown profile: https://aidanmarshall.ai/index.md - Markdown content map: https://aidanmarshall.ai/sitemap.md - Sitemap: https://aidanmarshall.ai/sitemap.xml