DZEKA
Engineer and solutions architect. I build systems end to end — Go, Rust, Kubernetes, in the open with every decision written down — and I have spent four years on the customer side of them: proof-of-concepts for tier-1 protocols, and AI tooling that automated away most of my own job. Walk up and press start.
PLAYER STATS
- 6TIER-1 POCS ARCHITECTEDAAVE · CHAINLINK · UNISWAP · POLYGON · YEARN · MORPHO
- ~80%OF COMPANY ARRSOLE TECHNICAL OWNER · TENDERLY
- ~60%SUPPORT LOAD REMOVEDOPENAI-POWERED INTERNAL ASSISTANT
- ~40%TRIAGE TIME CUTAI AGENTS · MACOS · WINDOWS · LINUX
- 283DECISIONS RECORDEDADRS ACROSS FIVE OPEN SYSTEMS
- 4 YRSIN THE CUSTOMER SEATSUPPORT → ESCALATIONS → POCS
SELECT SYSTEM
ALL CABINETS →PLATFORM & AI TOOLING
Getting software to run inside an environment you do not own — someone else's cluster, someone else's constraints, someone else's security review. This is the day job. The two cabinets here are a pair on purpose: they answer the two questions of every proof-of-concept in the order they get asked — can this tool run here, and is it any good here. Portcall has shipped; Assay is still being built, in the open.
Portcall
A pre-deployment network check in TypeScript — one command that tells a stranger's network whether the tool will run there.
Assay
M2 closed: every trial now runs inside a per-task container with memory, CPU and process ceilings and no network interface at all, and a diff is scored by running it - the ground-truth adapter measured at 1.0 and the null adapter at 0.0 rather than asserted. The pinned-image re-mine came back 743 examined and 0 kept again, so the reach limit stands and the negative result gets published. M3 is underway: `assay run`, n trials per task per tool, with pass^n leading.
The AI tooling I ship at work stays behind the badge.
Four years of it. An internal assistant that took roughly 60% of the manual load off a support team, and a cross-platform desktop app of triage agents — macOS, Windows and Linux — that ingests customer-reported issues, diagnoses them and outlines the fix, cutting engineering triage time by about 40%. It lives inside an employer's product, so there is no repo to link and nothing to play here. The outcomes are in the XP log. The engineering underneath it is the same engineering as the cabinets on this floor, and those are mine, in the open, decisions and all.
READ THE XP LOG →PAYMENT RAILS
Money movement, where correctness is the entire product: idempotency, double-entry ledgers, replay protection, reorg safety. Built on crypto rails because that is the unforgiving version of the problem — the rail reorganises underneath you and the ledger still has to be right. Bosun leads the row because it is the same problem one layer down: its golden path is a payment stack, and standing that stack up inside a cluster you do not own is the step before any of it can be correct.
Bosun
A Kubernetes operator and internal platform in miniature, in Go — one object becomes a whole environment.
Payment Rail
A stablecoin payment orchestration rail in Go — payments-grade correctness over crypto rails.
Tollgate
An x402 facilitator and payment gateway in Rust — pay-per-call APIs, verified at the gate.