Underwater networking is vital for enabling collaboration between divers, vehicles, and sensors in marine exploration, monitoring, and emergency response. Yet achieving reliable communication in such dynamic, bandwidth constrained environments remains challenging. Acoustic and radio frequency technologies suffer from attenuation, latency, and hardware overhead, while optical wireless systems typically require specialized transceivers or strict alignment, limiting practicality in mobile underwater networks. To address these limitations, we present AquaLink, a QR code–driven Optical Camera Communication (OCC) framework that enables robust underwater messaging using commodity smartphones and tablets. At its core, AquaQR employs blue–green 2-bit color encoding, Low-Density Parity-Check (LDPC) error correction, and geometric augmentations tailored for optical stability in turbid waters. An auto-configuration module adapts parameters before transmission, and a lightweight enhancement pipeline ensures real-time robustness under diverse conditions. Field trials in pool, lake, and coastal environments achieve over 90% decoding success at 5 m and up to 2× higher throughput than prior QR-based systems. By eliminating specialized hardware, AquaLink provides a scalable, low cost foundation for underwater visual networking, supporting message exchange, peer interaction, and localized link formation.

Story point estimation is a key practice in Agile project management that assigns effort values to user stories, helping teams manage workloads effectively. Inaccurate story point estimation can lead to project delays, resource misallocation and budget overruns. This study introduces Story Point Estimation using Reinforced Transformers (SPERT), a novel model that integrates transformer-based embeddings with reinforcement learning (RL) to improve the accuracy of story point estimation. SPERT utilizes Bidirectional Encoder Representations from Transformers (BERT) embeddings, which capture the deep semantic relationships within user stories, while the RL component refines predictions dynamically based on project feedback. We evaluate SPERT across multiple Agile projects and benchmark its performance against state-of-the-art models, including SBERT-XG, LHC-SE, Deep-SE and TF-IDF-SE. Results demonstrate that SPERT outperforms these models in terms of Mean Absolute Error (MAE), Median Absolute Error (MdAE) and Standardized Accuracy (SA). Statistical analysis using Wilcoxon tests and A12 effect size confirms the significance of SPERT’s performance, highlighting its ability to generalize across diverse projects and improve estimation accuracy in Agile environments.