AI systems
Agents and tool-enabled systems grounded in real engineering context, with permissions, evaluation, and human control designed in.
Independent consulting · Toronto & remote
Build AI systems that engineers can trust in production.
I help engineering organizations design and build production-grade agents, operational automation, and the systems around them—grounded in real data, explicit permissions, and measurable workflows.
Experience includes senior engineering roles at Microsoft, Meta, PayPal, and Coinbase.
Where I help
Technical leverage, applied carefully.
Focused engagements for teams with an important engineering problem—not an extra backlog to outsource.
Security & reliability
Production systems made easier to understand, recover, and operate—without treating security as a separate checklist.
Engineering productivity
Infrastructure and automation that remove operational toil, improve feedback loops, and leave teams able to own what was built.
Selected experience
Built where operational details matter.
Evidence from engineering roles—not consulting-client claims.
30M+
Azure virtual machines
Software engineering work on the Azure Guest Agent, deployed across more than 30 million virtual machines.
80%
fewer database incidents
Platform scaling work at PayPal that contributed to an 80% reduction in database production incidents.
90%+
internal adoption
Production diagnostics automation adopted by more than 90% of an approximately 100-engineer organization.
500%
faster build and test cycles
CI/CD and integration-test automation that improved engineering feedback speed by approximately five times.
Representative engagements
Scoped around an outcome.
Every engagement is shaped around the system, constraints, and result—not a block of hours.
AI for engineering operations
Design and build controlled AI systems that work with the operational context engineers already depend on.
Engineering workflow automation
Remove repetitive operational work with the right mix of deterministic automation and carefully bounded AI.
Observability & telemetry systems
Improve signal quality, operating usefulness, and cost without turning telemetry into another platform teams have to fight.
Writing
Notes from production engineering.
Practical writing about AI systems, automation, reliability, and the tradeoffs that appear after the demo.
What makes an MCP agent production-grade?
The difficult part is not giving a model tools. It is designing the permissions, evidence, evaluation, and operating model around those tools.
Automating engineering toil with n8n—without creating new toil
Workflow tools create leverage when the process, failure behavior, ownership, and role of AI are designed before the canvas fills with nodes.
Adding observability to an infrastructure stack
Start with the decisions operators need to make, then design signals, ownership, and cost controls that remain useful under stress.
Start a conversation
Have an expensive engineering problem?
Tell me what you are dealing with. If it is within my area of experience, I will tell you whether I think I can materially help.