Research Explainers
9 min read
Topic
Research Explainers
Essays, notes, profiles, and operating models collected under this topic.
Start Here
A few useful paths through Research Explainers
Research Explainers
Agent Coding Costs Hide in Review, Not Generation
A study of ChatDev traces finds that agentic coding systems spend most tokens on review and repeated context passing, not initial code generation.
Research Explainers Agents Change Work by Lowering the Cost of ExecutionA Perplexity field study compares conversational search with autonomous agent execution and finds large time savings, lower dissatisfaction, and broader task scope.
Research Explainers Synthetic Consumers Work Better When They Talk FirstA consumer research paper finds that LLMs can match human purchase-intent surveys when they answer in free text before being mapped back to Likert ratings.
Related Hubs
Move sideways by problem area.
Research Explainers
10 min read
Agents Change Work by Lowering the Cost of Execution
Research Explainers
10 min read
Synthetic Consumers Work Better When They Talk First
Research Explainers
8 min read
AGI May Be a Phase, Not the Finish Line
Research Explainers
10 min read
Post-Training Is Where Models Learn Bad Habits
Research Explainers
8 min read
Agent Harnesses Can Learn From Their Own Failures
Research Explainers
8 min read
New Technologies Take Time to Reach the Productivity Stats
Research Explainers
6 min read