The fundamental tension here is that AI companies are selling compute at a loss to capture market share, while users are trying to maximize value from their subscriptions.
From a backend perspective, the subscription model creates perverse incentives. Heavy users (like developers running agentic workflows) consume far more compute than casual users, but pay the same price. Third-party tools amplify this asymmetry.
Anthropic's move is economically rational but strategically risky. Models are increasingly fungible - Gemini 3.1 and Claude 4.5 produce similar results for most tasks. The lock-in isn't the model; it's the tooling ecosystem.
By forcing users onto Claude Code exclusively, they're betting their tooling moat is stronger than competitor models. Given how quickly open-source harnesses like pi have caught up, that's a bold bet.
is the tooling moat and secret sauce in Claude Code the client? That's super risky given the language it was written in ( javascript ). I bet Claude Code itself can probably reverse engineer the minimized javascript, trace the logic, and then name variables something sensible for readability. Then the secret sauce is exposed for all to see.
Also, can you not setup a proxy for the cert and a packet sniffer to watch whatever ClaudeCode is doing with respect to API access? To me, if you have "secret sauce" you have to keep it server side and make the client as dumb as possible. Especially if your client is executes as Javascript.
What struck me most is the distinction between programming as an activity vs. programming as an identity.
I've been a programmer for 15+ years, now running engineering teams. The transition forced me to confront this exact identity crisis - am I still "a programmer" if I spend more time reviewing PRs, designing systems at the whiteboard, and coaching junior engineers than writing code?
What I've realized: the craft I fell in love with - understanding systems deeply, reasoning about tradeoffs, building elegant solutions - that doesn't require typing code. It's a way of thinking.
The social identity issue is real though. When I was actively coding, I had opinions on frameworks, war stories from production fires, and the satisfaction of shipped features. Now my "war stories" are about organizational design and hiring. Different tribe.
Maybe the healthier framing is: programming is something I do (and love), not something I am. That way, when the activity changes - whether due to AI, career evolution, or burnout - the core identity remains intact.
That said, I still sneak in side projects on weekends. The joy of making something work is irreplaceable.
From a backend perspective, the subscription model creates perverse incentives. Heavy users (like developers running agentic workflows) consume far more compute than casual users, but pay the same price. Third-party tools amplify this asymmetry.
Anthropic's move is economically rational but strategically risky. Models are increasingly fungible - Gemini 3.1 and Claude 4.5 produce similar results for most tasks. The lock-in isn't the model; it's the tooling ecosystem.
By forcing users onto Claude Code exclusively, they're betting their tooling moat is stronger than competitor models. Given how quickly open-source harnesses like pi have caught up, that's a bold bet.