The problem

We were shipping multi-agent AI features, and the only record of what happened in production was a Datadog trace: spans, tool calls, token counts. Legible to the engineer who wrote the agent, useless to the product manager or the subject-matter expert who could tell whether the answer was any good.

So agent quality was anecdotes. "It felt smarter after the prompt change" is not a metric, and the people best qualified to judge the output had no way to see it.

How it works

What I built

• It reads production traces out of Datadog and rebuilds the whole interaction: what the user asked, which agents ran, which tools they called, and what came back. Multi-agent and multi-turn, assembled into something you can read top to bottom instead of a span waterfall.
• The review interface lives in Backstage, the developer portal everyone already had open. You filter down to the interactions worth looking at, read them, annotate them, and score the agent's behavior.
• I built it for product managers and subject-matter experts, not just engineers. The people who know what a good answer looks like were the ones who couldn't see the traces.
• Dashboards roll up quality, reliability, latency, usage, and cost, so agent quality sits next to the operational numbers instead of in a separate deck.
• Scored interactions get curated into Datadog experiment datasets. Today's human judgments become tomorrow's regression suite.

The problem

250 engineers who needed infrastructure, and a platform team that was the bottleneck for all of it. Every new service meant a ticket, a queue, and a hand-rolled Terraform module that drifted the moment it merged.

What I built

• A company-wide Backstage developer portal, built end to end, including custom TypeScript plugins for the service catalog, docs, dashboards, and scaffolding.
• Around 12 golden-path scaffolders that provision real production infrastructure: EKS clusters, AWS account vending, Lambda and API Gateway via SAM, Bedrock Agent runtimes, Secrets Manager entries, multi-environment ECR pipelines, and Istio + ALB internal ingress.
• Three control planes, split on purpose. Terraform for the long-lived infrastructure, ACK for IAM roles so IRSA could be granted on the fly, and Crossplane for resources born only at cluster bootstrap, like the per-cluster S3 buckets backing Thanos.
• Every path arrives wired: IRSA, GitOps delivery, and DNS managed in Terraform. Nothing ships half-configured.

The problem

A big share of engineering time goes to the same low-value loop: a pipeline fails, someone reads the log, someone makes a four-line fix, someone opens an MR. Multiply by 250 engineers. Merge requests rot the same way, collecting conflicts, stale branches, red jobs, and review comments nobody circled back to.

What I built

• An internal AI engineering platform that carries work from a plain-English request, a Jira ticket, an incident, or a failing pipeline through to a tested, reviewable change. You reach it from the web, Slack, the command line, or from other AI tools over MCP.
• It finds the relevant repositories and ownership context from source, service ownership, Jira, GitLab, and Confluence, then routes to specialized agents for research, implementation, bug fixing, testing, CI investigation, operational triage, security analysis, compliance, or cloud-cost work.
• For code changes it provisions an isolated workspace, checks out the right branch, makes the change, runs the tests and validation, and opens a properly attributed GitLab merge request linked to the originating task. Tasks run async, so nobody sits watching a terminal.
• It keeps working after the first commit: monitoring pipelines, investigating failing jobs, telling real failures apart from known flakes, applying bounded fixes when authorized, and rerunning validation. QA evidence is pinned to the exact commit SHA, so a green run from an older revision can never be passed off as proof for newer code.
• An MR caretaker classifies every open merge request continuously, spotting conflicts, red pipelines, unresolved findings, stale work, and missing validation, then dispatches bounded remediation to keep it moving toward a human handoff. Approval and merge stay deliberate human decisions.
• Scheduled agents cover recurring operations: release monitoring, repository health, cost-anomaly review, security checks. On-call agents correlate alerts with service ownership, source code, deploys, and telemetry.
• The coding model underneath is swappable. The platform is the durable part: task state, workspaces, agent selection, tool access brokered over MCP with OAuth, credentials, retries, evidence, and audit.

The problem

A ~13 minute container build sits in front of every single change an engineer makes. It isn't one problem. It's a tax on 250 people several times a day, and the cloud bill was climbing alongside it.

What I built

• A remote BuildKit fleet with ECR registry-backed layer caching, which took builds from about 13 minutes to about 2.
• Two CI lanes across the core repos: a fast one for drafts, the full gate once an MR is marked ready. Feedback speed where it matters, rigor where it counts.
• Terraform moved off laptops and into GitLab CI org-wide, with automated testing and validation, plan on merge request, and apply on merge.
• Cloud spend down 20% and holding, through KubeCost-driven optimization and application right-sizing. Earlier I designed a GCP demand-side platform that ran inside its performance requirements at about 40% under the cost projection.