Books

AI-assisted code often looks finished while resting on a premise nobody checked. A practical, evidence-based method for developers, tech leads and architects to find system-level defects before they ship.

The model is one box. This book is about every other box: retries, contracts, evals, budgets, guardrails and agents for backend engineers shipping LLM features to production.

Retries, timeouts, webhooks and month-end reconciliation: the parts tutorials skip. A practical book for backend engineers who build or own payment systems, from cards to real-time rails like Pix, UPI and QRIS.

Understand Redis data structures, caching, persistence, replication, and reliable operations through clear visual explanations.

Understand microservices through clear diagrams of service boundaries, data ownership, communication, resilience, and operations.

Kill an AI agent mid-refund and the retry bills every model call again and refunds twice. This book crashes one agent on purpose and recovers it on DBOS, Temporal, Restate and Prefect, with the real output of every run.

Understand how identity, devices, policy, enforcement, and telemetry work together in Zero Trust. Clear diagrams trace access decisions and explain practical security architecture, migration, and operations.

Build a practical understanding of Model Context Protocol from the first local server to production operations. Visual explanations connect protocol contracts, hosts and clients, authorization, testing, reliability, observability, deployment, and recovery.

Understand how independent AI agents discover each other, exchange messages, and coordinate tasks. Clear diagrams introduce Agent2Agent architecture, interaction patterns, security, and reliable operation.

We will define the developer contract first, and we will then build the platform underneath it together, step by step, using decisions that hold up in regulated environments and in small startups alike. Along the way, we will provision clusters, wire delivery, publish a developer API, enforce policy and ship observability, and we will do all of it where you can see every command, every manifest and every failure. Nothing here stays theoretical, because platform problems only show their true shape where an abstraction meets a real cluster.

Building reliable distributed systems is harder than it looks. This book takes you from the fundamentals of message queues to the realities of distributed transactions, with working code, practical patterns and production lessons to guide you. Learn how to handle failures, maintain consistency and build systems that keep working when things go wrong.

Every nanosecond is a decision. Most trading systems lose thousands of them before the strategy even runs.Architecting HFT Systems in Modern C++ is the build manual for the systems that don’t. Starting from bare metal, you’ll construct a complete high-frequency trading pipeline, from packet arrival to order on the wire, with zero heap allocation and zero system calls on the hot path.You’ll learn how to:Tune Linux for deterministic, low-jitter execution with isolated cores, IRQ affinity, and tickless kernelsReplace malloc with arena allocators, pre-faulted pools, and clocks that never touch the kernelWire every component through lock-free SPSC queues using only the memory ordering you actually needBypass the kernel with DPDK, ef_vi, and io_uring, reading packets exactly where DMA placed themDecode ITCH, SBE, and FAST feeds at wire speed with branchless, compile-time-dispatched parsersBuild a cache-resident order book that updates in nanoseconds, not microsecondsRun strategies through a zero-virtual-call harness and enforce risk limits in a handful of instructionsProve it all with deterministic pcap replay, a simulated exchange, and automated latency regression detectionEvery chapter ends with a Latency and Determinism Ledger, a running p50/p99/p99.9 record that shows exactly where each nanosecond went and how it was won back.No toy examples. No hand-waving. Just the architecture, the code, and the measurements behind real tick-to-trade performance.Speed without determinism is luck. This book teaches you both.Short version (for ads or the subtitle line):Build a complete HFT pipeline in modern C++, from kernel bypass to order on the wire, with zero allocation, zero syscalls, and every nanosecond measured.

사내 LLM 게이트웨이를 설계하고 운영하는 실무서입니다. 키리스 인증과 6단계 RBAC, 토큰 쿼터·예산과 비용 통제, 엔터프라이즈 규모 MCP, 운영 런북까지 다룹니다. 영문판 The AI Gateway Playbook의 한국어판(PDF 75쪽)입니다.

A practitioner's playbook for building an in-house LLM gateway: keyless auth, six-tier RBAC, quotas and budgets, per-token cost control, MCP at enterprise scale, and the operations to run it all. About 32,000 words with diagrams, configs, and checklists.

Design reliable, secure, and evolvable APIs with visual explanations, practical HTTP and OpenAPI examples, and production troubleshooting. Follow the path from service boundaries and contracts to resilience, safe change, and operations.