Related Experiment Videos
Agentic AI-Assisted Coding Offers a Unique Opportunity to Instill Epistemic Grounding during Software Development
Magnus Palmblad1, Jared M Ragland2, Benjamin A Neely2
1Center for Proteomics and Metabolomics, Leiden University Medical Center, Leiden2300 RC, Netherlands.
Abstract:
The capabilities of AI-assisted coding are progressing at a breakneck speed. Chat-based vibe coding has evolved into fully fledged AI-assisted, agentic software development using agent scaffolds, where the human developer creates a plan that agentic AIs implement. One current trend is utilizing documents beyond this plan such as project- and method-scoped documents. Here, we propose GROUNDING.md, a community-governed, field-scoped epistemic grounding document, using mass spectrometry-based proteomics as an example. This explicit field-scoped document encodes Hard Constraints (non-negotiable validity invariants empirically required for scientific correctness) and Convention Parameters (community-agreed defaults). In this framework, Hard Constraints are intended to function as field-scoped validity constraints that take precedence over lower-priority context when properly loaded, while Convention Parameters capture community-agreed defaults. In practice, GROUNDING.md will empower a non-domain expert to generate code, tools, and software that have best practices baked in at the ground level, providing confidence to the software developer but also to those reviewing or using the final product. It seems easier to have agentic AIs adhere to guidelines than humans, and this opportunity allows organizations to develop epistemic grounding documents in such a way that keeps domain experts in the loop in a future of democratized generation of bespoke software solutions.
Related Concept Videos
Non-equilibrium in the Cell
Cognitivism
Previously dominated by behaviorism, which prioritized observable behaviors and largely ignored mental processes, psychology transformed in the 1950s. Cognitive psychologists argue that understanding how we think and process information is...
Purposive Learning
Information Processing Approach
The Anchoring-and-Adjustment Heuristic
Creative Thinking
Divergent thinking is the...