SoulSenseAI

Sangle, Shreeraj Santosh (2025) SoulSenseAI. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (4MB) | Preview

Abstract

SoulSenseAI is a text-based persona-based mental-health companion that provides essential and transparent scalable care and addresses global care gaps. The platform uses a ReactJS front end, FastAPI orchestrator and Claude 3.5 Sonnet LLM built around the concept of engineering velocity but powered by clinical discretion. A benchmark test recorded a mean reaction of 850 ms (43 per cent quicker than the typical LLM responsiveness) as sustaining crisis-time stability. Four non-overlapping personas-maintained use-case coherence > 85 %, and increased match engagement by 15 20 %. A rule explainability layer and a sentiment-intent detectors layer produced a 8.7/10 comprehension rating compared to a 4.8 industry standard with only 91.2 % of the distress signals being flagged. A module-based PostgreSQL SQLAlchemy store, secure API gateway and privacy mechanism of the text messaging channel allows fast scaling without any intrusive sensors. New ethical blueprint SoulSenseAI integrates high performance, CBT-aligned micro-interventions and auditable reasoning to provide a textbook case of how technology can scale the delivery of compassionate care to underserved populations across the globe.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Thomas, Lavish UNSPECIFIED
Subjects:	Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence P Language and Literature > P Philology. Linguistics > Computational linguistics. Natural language processing R Medicine > RA Public aspects of medicine > RA790 Mental Health
Divisions:	School of Computing > Master of Science in Artificial Intelligence
Depositing User:	Ciara O'Brien
Date Deposited:	04 Jun 2026 15:22
Last Modified:	04 Jun 2026 15:22
URI:	https://norma.ncirl.ie/id/eprint/9344

Actions (login required)

View Item