Notes on applied AI, agent harnesses, RAG and payment infrastructure, from a software architect in Hong Kong.https://maniksoin.com/en-ushttps://maniksoin.com/blog/agents-in-the-payment-flow/https://maniksoin.com/blog/agents-in-the-payment-flow/Money movement is largely irreversible, which makes payments one of the harder places to point an autonomous agent. Here is how I currently think about putting one there safely.Sat, 04 Jul 2026 00:00:00 GMThttps://maniksoin.com/blog/building-tools-your-agent-can-use/https://maniksoin.com/blog/building-tools-your-agent-can-use/An agent is only as capable as the tools you give it, and tools are where the real engineering — and the real risk — lives. How I build safe, focused MCP tools for payment operations.Sat, 20 Jun 2026 00:00:00 GMThttps://maniksoin.com/blog/trusting-code-nobody-understands/https://maniksoin.com/blog/trusting-code-nobody-understands/Agents produce code faster than anyone can read it, and the deepest cost is comprehension debt, not bugs: software that runs while no one understands it. Verification can let you trust code you never read; that is necessary, and it is not the same as understanding it. What property-based testing, simulation, and proof can buy, and what they can't.Sat, 06 Jun 2026 00:00:00 GMThttps://maniksoin.com/blog/five-claude-code-workflow-patterns/https://maniksoin.com/blog/five-claude-code-workflow-patterns/Five agentic coding patterns I keep reaching for — sequential, operator, split-and-merge, agent teams, and headless — plus the machinery (CLAUDE.md, subagents, slash commands, plan-first) that makes delegation repeatable and verifiable.Sat, 30 May 2026 00:00:00 GMThttps://maniksoin.com/blog/from-rag-to-agent-harnessing-ordinance/https://maniksoin.com/blog/from-rag-to-agent-harnessing-ordinance/Why a one-shot RAG pipeline hit its ceiling on real regulatory questions, and how rebuilding Ordinance as an agent put the engineering in the harness around the model.Tue, 12 May 2026 00:00:00 GMThttps://maniksoin.com/blog/anatomy-of-an-agent-harness/https://maniksoin.com/blog/anatomy-of-an-agent-harness/The model is just one input into a running agent. The harness around it — prompt, tools, context, subagents — is the engineering discipline that decides whether an agent ships.Tue, 21 Apr 2026 00:00:00 GMThttps://maniksoin.com/blog/building-rag-for-hk-regulations/https://maniksoin.com/blog/building-rag-for-hk-regulations/How I built Ordinance — a retrieval-augmented generation system that answers questions about HK compliance with cited, verifiable responses across 224 government documents.Tue, 17 Mar 2026 00:00:00 GMThttps://maniksoin.com/blog/hello-world/https://maniksoin.com/blog/hello-world/Welcome to my blog. I'll be writing about software engineering, payments, and building products.Wed, 25 Dec 2024 00:00:00 GMT