Writing / Field notes

Ideas worked through in public.

Practical notes on software engineering, systems, and building things that hold up.

Systems / C++ / Rust

C++ for Mastery, Rust for Modern Delivery

What C++ has taught me about memory, why Rust works so well with coding agents, and what the recent DHH debate gets right.

Read article →
FERS / Systems / Research

FERS: From Research Simulator to Extensible Radar Platform

The lineage, architecture, engineering trade-offs, and evidence behind my master’s work on the Flexible Sensor Radar Simulator.

Read article →
Engineering / AI / Practice

Agentic Engineering Without Surrendering Understanding

How to use coding agents for speed while retaining ownership of architecture, invariants, failure paths, and the code that ships.

Read article →
Architecture / Deployment / Reliability

Designing a Deployment Pipeline as a System

A release pipeline needs input identity, explicit states, rollout evidence, rollback rules, and an honest success verdict.

Read article →
Architecture / Capacity / Performance

Capacity Is a Chain of Budgets

Why traffic, application instances, cache, database work, connections, and queues must be planned as one coupled system.

Read article →
Architecture / Deployments / Trade-offs

How Many Deployables Should a System Have?

How to choose service boundaries using resource needs, blast radius, release cadence, and maintenance cost.

Read article →
Architecture / Data / Consistency

The Source of Truth Is a System Design Decision

A practical way to decide which component may answer authoritatively when databases, caches, queues, and indexes disagree.

Read article →
Systems / C / Opinion

C Is God-Tier

Why C remains my favourite language for understanding the machine and building the layers above it.

Read article →
Architecture / Backend / Reliability

Stop Hiding Critical Work in Request Threads

A practical pattern for moving fragile background processing out of web requests and into durable, observable job systems.

Read article →
AI / Observability / Reliability

Use LLMs After You Have Already Done the Hard Part

A production log digest pattern where deterministic filtering, grouping, ranking, and evidence capture happen before any model call.

Read article →
General / Next.js / MDX

Hello World

Welcome to my new blog! This is the first post where I test out the MDX, LaTeX features, and the new search filtering system.

Read article →