Posts tagged: inference

Token Price Collapse and White-Collar Work — The iPhone Moment for Knowledge Labor

Frontier reasoning drops 10-100x by 2029. What happens to analysts, lawyers, accountants, engineers? Detailed phase-by-phase extrapolation with demand elasticity analysis.

Will LLM Tokens Become Like Bandwidth — Abundant and Cheap?

Mapping the submarine cable story to LLM inference — competition, open-source, and whether tokens will ever be "unlimited" like bandwidth.

The Future Stack — UI, Inference, Database

What if the traditional backend (REST APIs, business logic, middleware) gets absorbed by inference, leaving only UI, an LLM, and a database?

Self-Hosting DeepSeek V4 for Uncensored Local Inference

A practical guide to running DeepSeek V4 locally — model weights, hardware requirements, deployment options, and what you get vs the censored API.