Current Research Question: How do intelligent systems--biological or artificial--learn, reason, and build under uncertainty? Every system I build--from robotics and autonomous systems to credibility-based software and cognitive science--is another attempt to answer this question.
My projects begin with questions that remain difficult even after the first answer.
How do we justify believing internal representations exist?
How does intelligence forms in intelligent systems: Artificial or Biological?
How do intelligent systems update those representations from evidence?
How do we infer learning when learning itself cannot be directly observed?
How should uncertain behavioral evidence update confidence?
When is available evidence sufficient to justify building despite uncertainty?
Long-running projects where engineering, software, research, and uncertainty meet.
An immigrant-support platform designed to make information, housing, community knowledge, and practical resources easier to find, understand, and trust.
A public engineering and systems-thinking lab where I explore mechanical systems, robotics, CAD, software, and technical ideas through first-principles reasoning and documented building.
A credibility-first short-form learning platform exploring how technical content can be ranked using behavioral evidence rather than popularity alone.
Questions developed through literature, experimentation, system modeling, and technical inquiry.
What if software acknowledged you? Most interfaces silently execute commands. TalkUI tests a different interaction model: the interface briefly communicates what is happening before, during, or after an action. https://talkui.research-prototype.nirajchaurasiya.com/
The traces left by users, whether conscious or unconscious, may differ across domains and must be measured probabilistically with care.
A completed academic investigation using EPA air-quality data to examine the difference between observing environmental conditions through monitoring and extending those observations through modeling.
Working models for reasoning about learning, systems, evidence, and sufficient understanding.
An epistemic decision framework for determining when understanding is sufficient to begin acting, building, testing, or deciding under uncertainty.
A systems-thinking framework for defining a system, tracing its inputs and governing interactions, identifying its outputs, exposing assumptions and constraints, and preserving latent uncertainty before analysis or design begins.
An evidential interpretation framework for organizing observable behaviors according to how strongly they may support the inference that learning has occurred.
Essays and research notes documenting how the questions and systems are evolving.
An initial exploration of why humans may experience regret after simulating a better alternative, while an artificial system may perform a similar counterfactual update without an established subjective feeling.
An investigation into how behavioral signals should influence credibility-oriented ranking when each signal differs in intentionality, ambiguity, context, and evidential strength.
An investigation into why views, likes, retention, and virality should not be treated as direct evidence of credibility—and which behavioral signals might cautiously reduce uncertainty about a technical video’s perceived usefulness.
Videos, presentations, talks, demonstrations, and public explanations.
A public engineering series documenting the progressive design of a robotic hand through SolidWorks, mechanical reasoning, electronics, and control systems.
A short-form series questioning how someone can know that learning occurred rather than merely assuming it from attention, completion, or familiarity.
A developing talk examining why the feeling of learning can be mistaken for evidence that learning has actually occurred.
What I am currently building, researching, reading, preparing, and trying to understand.
Reach out about engineering, research, learning systems, speaking, collaboration, or an idea connected to the work.