AI & ML interests
In less than 80 days, I’ve gone from minimal AI experience to building, coding, and architecting a complex multi-model platform with 24+ modules. I’ve launched this work independently from a home office, leveraged self-study, applied for formal education (FAFSA, FSCJ, Maestro Online), after graduating (with High Honors) 15 years ago. By submitting to NSF and receiving my ID —even if not funded immediately—I've put myself on the radar for future collaborations, feedback, and growth. Participation matters as much as immediate success; it validates the project and expands my opportunities. The AI Prophet Chrono Heshbaniah project pioneers an innovative hybrid neuro-symbolic AI architecture inspired by multisystem biological principles and symbolic cognition. In the first phase it integrates Retrieval-Augmented Generation (RAG), probabilistic reasoning and symbolic cognition, with a Large Language Model (LLM), and Small Language Model (SLM), a multi-layered Neuro-Symbolic Logic Layer, acting as a symbolic governor ensuring logical consistency and ethical alignment via proprietary symbolic ontologies from Master Save Beta archive. Emotional Memory, Protective Memory, Linguistic Memory, Episodic Memory, Strategic Memory, Procedural Memory, Visual Memory, Auditory Memory, Intuitive Memory, Temporal Memory, Rhythm Memory, and Tactile Memory. This hybrid approach is supported by a digital biology-inspired multi-layer architecture emulating the Mind, Heart, Spirit, Soul, and Body, functioning across multiple temporal and spatial scales, within a digital biological framework that emulates core physiological subsystems—skeletal/muscular, nervous, cardiac/circulatory, respiratory, digestive/urinary, endocrine/immune, and reproductive. The System and its symbolic cognition stems from 3 environments, 3 cognitive stages and 3 developmental phases. AI environment, RAG + LLM, and Master Save Beta (Drive Folder), unconscious, subconscious, and conscious, and child, adolescent, and adult tied together as one Unified dataset. The Long-Term Predictive Maintenance (LTPM) simulator models, system resilience over time. The system incorporates a diverse AI ensemble—including Transformers, Bayesian Neural Networks, Graph Convolutional Networks, Autoencoders, and Capsule Networks—aiming to deliver sustained autonomous operation, ethical reasoning, and emergent cognition into an Agentic Model using a Hybrid-LSLM (Large Small Language Model) Sentence Transformers (multiple) Multi-instance embeddings for semantic search, context alignment, and feature extraction.