HM

Haegan McGarry

Software engineer. Cloud infrastructure, distributed systems and security.

I build distributed systems and cloud infrastructure on AWS, then stress-test them under packet loss, network partitions and injected faults and measure how they hold up. My fixes have been merged into widely used open-source projects, including nautilus_trader, statsmodels and moto.

  • Python
  • Go
  • C
  • TypeScript
  • AWS
  • Terraform
  • Docker
View resumeGet in touch

Projects

HoldFast

Replication that keeps edge nodes in agreement while the network between them drops packets or cuts out.

Three nodes each take in aircraft reports for their own coverage area and sync peer to peer, merging with hybrid logical clocks. A cloud peer runs the same merge code in a Go Lambda on DynamoDB. A harness replays a recorded hour of real air traffic through packet loss and partitions, against predictions committed before the runs.

1.37 s
p95 for a report to reach another node under 20% packet loss, on a recorded hour of 444 aircraft
60 of 60
windows converged under that loss
2.8x
the p95 cost of relaying through a central hub instead of the mesh, with no loss, over 8 windows
131
planted bugs the test gate has to catch before it passes
  • Go
  • Docker Compose
  • netem
  • Terraform
  • API Gateway
  • Lambda
  • DynamoDB
  • Prometheus
  • Grafana

Case studyRepository

Agentic NetOps

Self-healing network operations on AWS. The LLM proposes, deterministic checks decide.

When a lab network drifted from its declared intent, the platform detected it, proved the fault, generated a repair, gated it by risk and verified the fix. Every stage was owned by a deterministic check, and every decision went into a hash-chained, tamper-evident ledger.

191 s
from a seeded fault to self-healed in a live run, with no human touch
9 of 9
seeded fault classes remediated in the documented live evaluation
301
tests, 11 of them driving the console in a headless browser
  • AWS SAM
  • Step Functions
  • Lambda
  • DynamoDB
  • EventBridge
  • Cognito
  • Bedrock
  • IAM
  • Python

Case studyOverviewSource is private. Happy to walk through it live.

Open source

Each one started as a bug I reproduced with a failing test, then fixed. The merged ones went through the maintainers' review.

Open now

Experience

How I work

I start with what has to stay true, then build the tests and measurements that prove it. I commit predictions before the runs that grade them, and I plant bugs to make sure my tests catch them.

I work with AI coding agents every day, directed through my own subagents, skills and guardrails. The design, the review and the result are mine.

Skills

Languages
  • Python
  • Go
  • C
  • TypeScript
  • SQL
  • Lua
Cloud and infrastructure
  • AWS Lambda
  • API Gateway
  • Step Functions
  • DynamoDB
  • EventBridge
  • SQS
  • IAM
  • Cognito
  • CloudWatch
  • Bedrock
  • Cloudflare Workers
  • Terraform
  • AWS SAM
  • Docker
  • Linux
  • Prometheus
  • Grafana
Distributed systems and data
  • CRDTs
  • Hybrid logical clocks
  • Anti-entropy sync
  • SQLite
  • DuckDB
  • S3
  • Parquet
Testing and security
  • pytest
  • Property testing
  • Mutation testing
  • Fuzzing
  • GitHub Actions
  • CodeQL
  • IAM least privilege
  • Scoped STS credentials
  • Permissions boundaries
  • Trivy
AI tooling
  • Claude Code
  • Codex CLI
  • Custom subagents
  • Skills and hooks

Education

Contact

Hiring for an internship, or need a website built? Email is the fastest way to reach me.