Biotechnology Journal International

Crop improvement increasingly depends on biotechnological methods that can accelerate genetic gain while preserving yield under drought, salinity, heat, nutrient limitation and biotic stress. Yet the field is often discussed as a collection of technologies rather than as a translation problem linking target discovery, genetic intervention, plant regeneration, field validation and breeding deployment. This critical narrative review evaluates the principal biotechnology platforms used to improve crop yield and stress tolerance, with emphasis on evidence maturity, field relevance, genetic architecture and the causes of inconsistent translation. Literature published primarily from 1 January 2015 to 8 August 2026 was examined, while earlier seminal studies were retained when they established concepts or field-tested traits that remain influential. The synthesis covers genomics-assisted selection, pan-genomics and phenomics, transgenic and RNA-based engineering, CRISPR-associated genome editing, base and prime editing, synthetic and metabolic engineering, and transformation and regeneration technologies. Evidence is strongest where biotechnology is integrated with breeding and tested against agronomically meaningful outcomes across realistic environments. Genomic selection is particularly suited to polygenic traits, but prediction decays when training populations poorly represent target environments. Transgenic and genome-edited interventions have generated convincing examples of improved drought performance, salinity tolerance and disease resistance, although effect sizes are frequently context dependent. Cis-regulatory editing and allelic series offer a promising route to tune quantitative traits while reducing pleiotropic penalties, whereas base and prime editing broaden sequence precision but remain constrained by delivery, editing efficiency and crop-specific regeneration. Photosynthetic and metabolic engineering has produced notable field proof-of-concept, yet transfer from model or experimental systems to major crops remains incomplete. Across platforms, the dominant bottlenecks are genotype-dependent transformation, genotype-by-environment interaction, combined stresses, fitness trade-offs, durability of resistance, and insufficient multi-environment validation. The most defensible strategy is therefore not technological substitution but an integrated pipeline in which genomic prediction and phenotyping identify robust targets, precision engineering generates controlled allelic variation, and multi-environment breeding determines whether molecular gains become stable yield advantage.