everyone would have a deeper appreciation for Agent Products that rock because of great Context/Harness Engineering if they… talked to: - LL…

everyone would have a deeper appreciation for Agent Products that rock because of great Context/Harness Engineering if they… talked to: - LLM Base models - Post-Trained models with no harness (no tools, no built in prompts, nothing) helps internalize how much stuff needs to happen to shape a massive blob of imperfect intelligence to rock at your task from Post-Training to Harness Eng building a great Agent Experience is hard because building a great product is hard sweating the details matters and every seemingly small change can have an outsized impact on the overall experience (instructions, skills, model choice). It helps to have some system to measure these changes as you evolve your product (agent) over time talk to a base model! it’s fun and gives you an appreciation for how much work goes into that “agent that just works”

Source: Harrison Chase (X) | 2026-05-05