Blog
Writing
Notes on backend and AI engineering, security tooling, and homelab networking. Originally posted on LinkedIn (opens in a new tab).
· 1 min read
Cutting code review noise with Jev, a millisecond decision model
Adding Jev, a non-autoregressive decision model, as the final gate in a 3-tier automated PR review pipeline to filter out low-signal findings.
· 1 min read
How I reach my homelab database from anywhere, without exposing it to the internet
How I reach Postgres on my homelab from anywhere with Tailscale and PgBouncer, without opening a single port to the internet.
· 1 min read
Your homelab doesn't need port forwarding. It needs Tailscale.
Tailscale gives a homelab private, encrypted remote access without a public IP or port forwarding, and it's a low-risk way to learn real networking.
· 2 min read
Testing Strix, part 3: what held up under manual review
Manually verifying every Strix finding: which ones were real, which were inflated or wrong, and why human review still matters.
· 2 min read
Testing Strix, part 2: black-box vs white-box scans
Running Strix black-box against a live app and white-box against its source produced very different findings. The live run was far stronger.
· 1 min read
Testing Strix, part 1: a pentester agent, not a linter
Why I tried Strix, an open-source LLM agent that attempts real exploits to prove its security findings instead of just flagging suspicious code.
· 1 min read
Building Smarter with OpenCode + NVIDIA: Here's What I've Learned
How I use OpenCode with NVIDIA NIM's free models, giving each model the job it does best: audit, review, or code generation.