Manager · Platform EngineeringResMed · since 2021Halifax, Nova Scotia · AT

Building platforms that scale.— Raghav Gupta, Manager, Platform Engineering at ResMed

I'm Raghav — I lead package and observability engineering at ResMed, working across AWS cloud infrastructure, the GitHub build platform, and developer experience for 20+ teams. I also build AI agents — a full agent factory — and run my own Proxmox homelab.

In one breath

Manager, Platform Engineering, ResMed.

  • Leads the Package & Observability function at ResMed
  • Associate developer → Manager in four and a half years (Jul 2021 → Dec 2025)
  • Builds the platforms 20+ engineering teams ship on — AWS, GitHub builds, observability
  • Ships with AI agents: this site is built by a two-agent factory behind CI gates

§01 — Expertise

What I work on.— cloud & platform engineering, build, packaging & release, observability, ai agents & agent factory, homelab & self-hosting, leadership & collaboration

AWS and the cloud, the GitHub build and packaging platform, observability — and the AI agents and homelab where the next ideas get proven. The areas I work across, and the tools I reach for.

01

Cloud & Platform Engineering

Cloud-native infrastructure on AWS that scales — from Kubernetes to Terraform modules — designed to be reliable, cost-efficient, and a pleasure to build on.

AWS · EKS · LambdaTerraform / IaCCost optimization
02

Build, Packaging & Release

The build platform on GitHub: fast, reproducible pipelines, the packages and artifacts they produce, and the release flow that ships them with confidence.

GitHub ActionsSelf-hosted runners (ARC)Package managementReusable workflows
03

Observability

You can't improve what you can't measure. I lead the observability function — metrics, logs, traces, and alerts that give teams real visibility and cut MTTR.

Datadog migrationSLIs / SLOsOpenTelemetry
04

AI Agents & Agent Factory

I build AI agents and the factory that produces them — agents that plan, code, review, and ship behind the same CI gates as people, wired to tools through MCP.

AI agent factoryMulti-agent workflowsMCP servers & toolsLLM apps & RAG
05

Homelab & Self-hosting

A Proxmox homelab I run myself — the proving ground where new infrastructure, runners, and agents get built and broken before they reach anything that matters.

Proxmox VEVirtualizationSelf-hosted services
06

Leadership & Collaboration

Technology is a team sport. I lead the package and observability function, mentor engineers, and align platform decisions with the goals of the business.

Team leadershipStakeholder mgmtMentorship & hiring

§02 — Experience

A track record of building.— Manager, Platform Engineering and DevOps experience at ResMed, Waterford Energy Services, Dalhousie University

Platform work at ResMed — from foundational IaC to leading the function — built on a full-stack and DevOps foundation.

ResMed

2021 — Present
Halifax, NS · 5 yrs
Manager, Platform EngineeringDec 2025 — Present
Senior Engineer, Platform EngineeringDec 2023 — Dec 2025
Software Engineer, PlatformJul 2022 — Dec 2023
Associate Software Developer, Platform · AWSJul 2021 — Jul 2022

Building the tooling and platforms that automate configuration and deployment of cloud products — reliable, secure, observable systems — across 20+ engineering teams. Grew from authoring foundational IaC to leading the Package & Observability function, with deep work on the GitHub build platform and AWS.

Waterford Energy Services

2019 — 2021
Halifax, NS · 2 yrs 4 mos
DevOps Lead / Analyst / DeveloperAug 2019 — Jul 2021
Software Developer · IT AnalystAug 2019 — Jul 2021
Software Developer / Analyst · InternshipApr 2019 — Aug 2019

Full-stack development on ASP.NET MVC 5 — Razor, HTML/CSS, jQuery — within a Git and Scrum workflow. Led DevOps: automating and monitoring deployments to staging and production across applications.

Dalhousie University

2018 — 2019
Halifax, NS · 1 yr 8 mos
Student IT Navigator · IT Support SpecialistMay 2018 — Dec 2019
Teaching AssistantMay 2019 — Aug 2019

Campus IT for Information Technology Services — network support, hardware/software troubleshooting, and ticketing — alongside teaching assistant work.

§03 — Lab

Proof, not claims.

Three working pieces of the job — break one, question one, inspect one.

A small production system, simulated live in your browser. Break it — spike the traffic, kill pods, slow the database, ship a bad release — and watch the safeguards a platform team builds catch it. Switch one off to see what it was quietly preventing.

Traffic flows from users through the Cloudflare edge and an API gateway to a pool of service pods backed by a database. usersedgegatewayservice podsdatabase

All systems normal

    Type help for commands, or ask a question in plain English. Where your browser ships an on-device model it answers locally, grounded only on this profile — nothing is sent anywhere.

    This site is the portfolio. How it was built — agents write the code, CI decides, a human ships — and how it reached you, measured live for this visit.

    How it was built

    1. Agent claims the task.agents/TASKS.md — Claude Code or Codex, never both on one file
    2. Plan → code on a branchnothing lands on main without a PR
    3. LintESLint strict + type-checked · 0 errors allowed
    4. FormatPrettier check
    5. Type-checkastro check · TypeScript strict
    6. BuildAstro SSR bundle for Cloudflare Workers
    7. Runtime smokeboots the built Worker in workerd and hits every surface
    8. Human review & mergepull request on GitHub
    9. Deploymanual workflow_dispatch · re-gated · wrangler deploy
    10. AnnounceIndexNow ping → answer engines recrawl

    Runs on setkernel-linux — self-hosted, autoscaling GitHub Actions runners (ARC). This page: build .

    How it reached you

    1. You → edgemeasuring…DNS · TLS · first byte, from Navigation Timing
    2. Edgemeasuring…the Cloudflare data centre that answered you
    3. Workermeasuring…Astro SSR on Workers, edge-cached per deploy
    4. Agentsmeasuring…live MCP 2026-07-28 round-trip to /mcp
    5. In your browsermeasuring…WebMCP tools registered for on-page AI agents

    §04 — Quick answers

    The short version.

    Who is Raghav Gupta?

    Raghav Gupta is Manager, Platform Engineering at ResMed, based in Halifax, Nova Scotia, Canada, and has been at ResMed since 2021. Building platforms that scale.

    What does Raghav specialize in?

    Cloud & Platform Engineering, Build, Packaging & Release, Observability, AI Agents & Agent Factory, Homelab & Self-hosting, and Leadership & Collaboration. Day to day that means AWS · EKS · Lambda, GitHub Actions, and Datadog migration.

    Where has Raghav worked?

    ResMed (2021–Present), Waterford Energy Services (2019–2021), and Dalhousie University (2018–2019). At ResMed Raghav progressed from Associate Software Developer, Platform · AWS to Manager, Platform Engineering.

    What kind of work can I bring to Raghav?

    Build & packaging platforms, Cloud infrastructure on AWS, Observability & reliability, AI agents & agent factories, Internal developer platforms, and Platform leadership. Based in Halifax, Nova Scotia, Canada (America/Halifax) and reachable by email from anywhere: hello@raghavgupta.ca.

    How can I contact Raghav?

    Email hello@raghavgupta.ca — it's the best way to reach Raghav. Also on GitHub (https://github.com/realraghavgupta), and LinkedIn (https://www.linkedin.com/in/realraghavgupta/).

    Can AI agents read this profile?

    Yes. There is a public, read-only MCP server at https://raghavgupta.ca/mcp, plus https://raghavgupta.ca/llms.txt, https://raghavgupta.ca/llms-full.txt, and schema.org JSON-LD at https://raghavgupta.ca/profile.json. Requesting the homepage with Accept: text/markdown returns Markdown.

    §05 — Contact

    Let's talk. — contact Raghav Gupta

    Hiring, a hard problem, or a second opinion — email is the fastest way to reach me.