SH.Consulting
Insights

Insights

Perspectives on AX, AI agents, and harness engineering. Canonical long-form versions of posts published on LinkedIn.

2026-06-15

Who should build AX for it to succeed?

AX (AI transformation) succeeds when it is built by the frontline people who do the work every day, not by outside developers. The exceptions and tacit rules hidden in a workflow are known only to the person who runs it daily. So the essence of AX education is not adding more developers, but enabling frontline workers to automate their own work through vibe coding. A finance and operations staffer who automated their daily ordering process shows what this looks like.

Read more →
2026-06-14

Why is automating your own daily life the best starting point for AX?

Removing one small task you repeat every day — entirely, not partially — is the best starting point for AX (AI transformation). The reason is that taking a task fully out of your own hands, rather than making it a few percent faster, is how you learn firsthand what AI can replace and how far you can trust it. A personal example: AI picks schedule items out of scattered KakaoTalk chats, registers them on a calendar, and delivers a single morning briefing.

Read more →
2026-06-13

Why Should You Get AI Output as HTML Instead of Markdown?

For finished deliverables, receiving AI output as HTML instead of markdown dramatically speeds up comprehension. AI now produces documents faster than people can read them, and markdown beyond 100 lines effectively goes unread. An Anthropic engineer on the Claude Code team made the same call — "HTML is the new markdown" — and started receiving output as HTML. When you need interactive views, the Playground plugin closes the gap. The rule is simple: finished artifacts you need to grasp structurally go to HTML; intermediate artifacts stay in text.

Read more →
2026-06-13

What decides AI coding productivity — the model or the workflow?

The quality of AI-assisted coding is decided more by workflow than by model choice. The real bottleneck is rarely a weak model; it is vague specification and unfinished work. oh-my-claudecode (OMC) is an orchestration layer on Claude Code that hardens three habits into tools: deep-interview clarifies what to build, ralplan validates how to build it, and ralph guarantees it actually gets finished. The point is turning "what, how, and to the end" into an enforced pipeline.

Read more →
2026-06-12

Is the Contact Center a Cost Department or a Source of New Business?

A contact center is not only a cost department; it can be a source for discovering new services and lines of business. Incoming customer inquiries already contain clues to unmet demand, and turning call records into assets with STT and an LLM wiki lets you mine those clues systematically. The remaining hurdle is masking personal data — solve it, and the contact center turns from a cost department into a business-discovery engine.

Read more →
2026-06-11

Why Does Frontend Knowledge Still Matter When AI Writes the Code?

Frontend knowledge is the map that determines the level of work you can delegate to AI. Terms like React, npm, build, and SSR are not items to memorize but a chain of problem-solving; once you know what problem each technology was born to solve, the language of your instructions changes, and the same AI produces different results. This is why vibe coding education should cover the evolution of the technology before tool usage.

Read more →
2026-06-11

What Is a Backend? A Genealogy of Problem-Solving That Began with a 1993 Guestbook

A backend is the part of a service where a server receives requests, processes data, and returns results — and it was born in 1993 while solving the guestbook problem. Pre-built HTML files could not show the next visitor a message someone had just left, so servers began assembling pages on the spot for every request; that was the starting point. The thirty years of evolution that followed repeat the same pattern: a problem appears, and a technology emerges to solve it. Once non-developers know this genealogy, they can give AI concrete instructions about the backend.

Read more →
2026-06-11

Why Does AI Keep Recommending Supabase in Vibe Coding?

AI keeps recommending Supabase during vibe coding because two technology streams that originally evolved separately, databases and storage, converged into this single platform. Supabase is an open-source platform that puts a proper PostgreSQL under Firebase-style convenience and bundles authentication and storage with it, so things you once had to set up and wire together separately now come connected from the start. Once you know this lineage, the quality of the instructions you give AI changes.

Read more →
2026-06-11

What Is Deployment — and Why Won't localhost Open on a Friend's Computer?

Deployment is the process of putting code that exists only on your own computer onto a server so that anyone on the internet can use it. This article covers why a vibe-coded app only runs on localhost:3000, how deployment evolved from FTP to Git, Docker, and CI/CD, and what platforms like Vercel do for you — and at what cost. With this map in your head, the questions you ask AI change, and that specificity is what separates skill levels in non-developer vibe coding.

Read more →
2026-06-11

Why Does AI Output Probabilities Instead of Rules?

Today's AI is not a machine that executes rules written by humans; it is a machine that outputs probabilities, such as "93% likely spam," based on patterns it finds in data on its own. Problems with clear conditions, like bank transfers, are solved perfectly by rule-based code, but rules collapse on problems with many exceptions and blurry boundaries, like image recognition. Machine learning and deep learning emerged from this failure of rule-based systems, and this distinction becomes the first criterion for decomposing work in AX consulting.

Read more →
2026-06-11

Why Doesn't ChatGPT Remember Your Conversations?

ChatGPT and other LLMs do not remember conversations. Each turn, the entire previous dialogue is sent again, and the model reads that bundle like text it has never seen before, simply predicting the next token. This is why the context window fills up and costs balloon as conversations grow longer. From the tokenizer's inefficiency with Korean to hallucination, temperature, and system prompts, all of it can be verified on screen with just a few lines of code.

Read more →
2026-06-11

What Is AI Orchestration? The Design That Begins Where RAG Chatbots Stop

AI orchestration is the design of control flow that lets AI handle complex real-world work: instead of a straight question-search-answer pipeline, it builds loops that backtrack when stuck and branches that change course with the situation. LangChain supplies the standard parts, LangGraph the control flow, and LangSmith the observability — only together do they form one system. Attaching them to a job a single API call can finish is wasteful; they are tools to bring out when the work is complex enough to need loops and branches.

Read more →
2026-06-10

What Is Harness Engineering?

Harness engineering is the practice of improving AI output quality not by changing the model itself, but by designing which tools the model can use and which rules and workflows it must operate within. The same model performs very differently depending on its harness — so before upgrading to a more expensive model, this is the layer to inspect first.

Read more →
2026-06-10

AI Native vs. AI Assisted — Which One Is Your Company?

AI Native means a structure where entire workflows complete through AI with almost no human intervention; AI Assisted means people keep doing their jobs as before while AI helps with parts on the side. Most corporate "AI Native declarations" are in fact AI Assisted — and companies that cannot tell the difference see productivity stay flat even after adopting AI.

Read more →
2026-06-10

Why AX Projects Fail — Don't Help the Team, Eliminate the Work

The most common failure pattern in AX (AI Transformation) projects: an AX team tours departments, collects requirements, and builds tools — which the field then never uses. Successful AX starts not from "help that team" but from "eliminate that team's unit of work entirely," and the people are not cut but transitioned to new roles.

Read more →
2026-06-10

Generative UI and MCP Apps — Where Are Agent-Era Interfaces Headed?

Agent UI is evolving through three stages: static (agents fill data into pre-built components), declarative (agents write a JSON description that a rendering engine maps onto a design system), and generative (the model writes HTML/CSS at runtime). Today's equilibrium is declarative; the gate to generative is not capability but trust — a safe distribution structure, i.e., sandboxing. MCP apps are drawing attention as that distribution channel.

Read more →
2026-06-10

What Is K Public Data MCP? — A Server That Connects Korean Public Data to AI

K Public Data MCP is an open-source MCP (Model Context Protocol) server that unifies South Korea's scattered public data APIs — the Ministry of Government Legislation's law database, the DART corporate disclosure system, and the public data portal — so that "search the Civil Act," "show Samsung's financials," or "find pharmacies in Gangnam" work directly in an AI chat. It runs on a single connector URL, with no API key issuance or installation required.

Read more →