Hi, I'm Harshith Sunku. From CPU registers to UI, cloud, and AI. Happiest close to the metal. No language barrier. No stack barrier.
// THE WHOLE STACK
I've written bootloaders in Assembly, kernels and network stacks in C, firmware for tiny devices, APIs and web apps, infra tools, and AI integrations. Whatever the layer, I'll meet you there.
But if you ask where I'm happiest — it's the bottom three rungs: boot, kernel, and network. That's where the abstractions run out and the machine tells the truth.
More about me →♥ where I'm happiest
// START HERE
A dedicated container on my Proxmox host runs a coding agent that writes, reviews, and ships changes to this site — through the same git-push pipeline I use. The interesting engineering isn't the model; it's the guardrails around it.
No VPS, no database, no CMS bill. This site is a git repo that compiles to static files — search index, social cards, and fonts included — and Cloudflare Pages serves it for free. Here's the whole pipeline, with the actual configs.
// LATEST WRITING
VLANs decide which networks can talk. They say nothing about which people can. Part 5 of the homelab series: a central identity service in its own VM, Tailscale as the front door, and why the break-glass path matters more than the lock.
The plan was to consolidate everything onto one Proxmox box. Then a perf tool needed real PMU counters, a firmware flash needed a real serial port, and testing on ARM meant a phone running a custom kernel. Meet the fleet — and what each machine taught me.
Wake-on-LAN only works from inside the LAN — so I put a $5 ESP32 on the LAN. It subscribes to an MQTT topic and broadcasts magic packets on demand. The build, the failure modes, and what a hardware watchdog is really for.
The lab started as services sprawled across whatever machine was awake — until one bad upgrade took them all down at once. This is what that failure taught me: one Proxmox host, ZFS snapshots as the undo button, and a network segmented by trust.
A language model generates one token at a time, and each token re-reads gigabytes of weights and a growing cache. Understanding the KV cache and the memory-bandwidth wall explains latency, context limits, and every serving optimization from the last two years.
A flame graph turns thousands of stack samples into one picture where width is time and the widest boxes are your problem. How sampling profilers work, how to read the graph in ten seconds, and the traps — inlining, wait time, broken stacks — that mislead beginners.
// SERIES
SERIES · 1 PART
How this blog actually runs: the static build, the free hosting, the search index, the CI pipeline — every moving part, with real configs and real costs.
SERIES · 5 PARTS
The lab exists to learn on. Every post is the same shape: a real problem I hit, what I tried, what actually worked, and what it taught me — Proxmox and ZFS, a fleet of bare-metal misfits, an AI coworker, and the small devices that keep it all recoverable.
// THINGS I'VE BUILT
A software TCP/IP stack built entirely in C — Ethernet framing, ARP resolution, IP routing, and TCP state machine via Linux TAP devices. Validates by fetching real HTTP content over the custom stack.
Self-hosted SSH key manager for Proxmox — distributes public keys to LXC containers and QEMU VMs via the Proxmox API and pct/qm guest exec, with zero prior SSH access to the guest. Create a container, SSH in immediately.
A free, hands-on course on building AI agents with a fast brain and a slow brain — 12 short notebooks and 13 real use cases that put Jev, a decision model, inside an LLM agent loop. One API key; the whole course runs for under $0.20.
// HOW I WORK
Measure, don't guess
Opinions are cheap. Data, tests, and benchmarks are not.
Understand the layer below
You don't own your abstraction until you understand what it hides.
Rebuild it to learn it
Writing it from scratch is the fastest way to understand why the real one works.
Cargo-culting best practices
If you can't explain why you're doing it, you're not engineering — you're copying.
"Not my layer"
The bug doesn't care where your job description ends. Follow it wherever it goes.
"It works on my machine"
Reproduce it, isolate it, or it isn't fixed.
STILL SCROLLING?
Bootloaders, backends, or agent loops — I'm always up for a good engineering conversation.
