Subodh Jena
BlogNotes

Blog

Writing

Life, tech, and everything in-between.

ReAct: Reasoning and Acting in One LLM Loop

Mar 11, 2026·6 min·AI

The foundational agent pattern from Yao et al. interleaves a short reasoning trace with a single tool call, then folds the observation back into the next step. Simple in shape, strong on grounded reasoning benchmarks.

Workflows Versus Agents in LLM Systems

Feb 16, 2026·7 min·AI

The term agent has become overloaded, and the overloading is expensive. A system in which the language model controls its own flow costs more to run and more to debug. The default should be a workflow.

12

Writing

BlogNotes

Work

ExperimentsPortfolio

Connect

AboutContact

© 2026 Subodh Jena

X (Twitter)GitHubLinkedIn