Research & Deep Dives
Research papers, reports, scenarios, and technical findings translated into practical implications for builders and operators.
Open Research & Deep DivesFAPO uses Claude Code to optimize multi-step LLM pipelines by inspecting intermediate failures, trying prompt fixes first, and changing the workflow only when evidence points to a structural bottleneck. The result is a practical diagnostic loop for builders.
Michael Mauboussin and Dan Callahan use PVGO to separate current business value from future growth expectations, giving investors a cleaner view of what the market has already priced in. It works as an expectations lens, not a timing signal.
AI-native startups are smaller, flatter, more technical, and more valuable per employee because AI is embedded in the product, not merely internal workflows. The analysis connects organizational design, technical founder density, and valuation economics.
A detailed scenario about agents, automated AI research, model misalignment, geopolitical competition, and the choice between racing and slowing down. Its value lies in testing assumptions about timelines, coordination, and control rather than treating the scenario as destiny.
A study of ChatDev traces finds that agentic coding systems spend most tokens on review and repeated context passing, not initial code generation. The finding shifts optimization toward handoffs, critique loops, and context efficiency.
A Perplexity field study compares conversational search with autonomous agent execution and finds large time savings, lower dissatisfaction, and broader task scope. The comparison shows why autonomy changes both the economics and ambition of knowledge work.
A 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. The result favors better survey design over treating synthetic respondents as wholesale human replacements.
A DeepMind report argues that human-level AGI may be followed by several pathways toward superintelligence, each constrained by data, compute, embodiment, and coordination bottlenecks. It reframes AGI as an intermediate capability threshold rather than a stable endpoint.