I'm a software engineer building AI systems at Tally Solutions, working across the AI systems stack, from agentic frameworks and retrieval to evaluation, observability, and the infrastructure that makes these systems reliable at scale.
Outside work, I explore assistive AI that can make technology more useful without losing the meaning, individuality, and human expression behind how people communicate.
Developing enterprise GenAI applications for product support and software engineering. Contributions span context-aware retrieval, agent workflows, model evaluation, and inference optimisation, alongside shared infrastructure for memory, safety, recovery, and observability.
Worked on computer vision for inventory workflows, combining edge-model optimisation with CLIP-based classification. Quantised and benchmarked vision models and developed a pipeline for identifying industrial chemical labels.
Assistive speech recognition and communication that keeps literal alternatives separate from editable messages, with profile-scoped context and explicit controls over speech delivery.
View repositoryA source-aware vocabulary application with FSRS scheduling, durable review synchronization, and recoverable document ingestion. Experimental semantic assessment explores how typed answers can be evaluated against intended meanings.
View repositoryA Jetson and LEGO EV3 robot combining person recognition, active camera control, and route-checked movement. Identity continuity and state-bound inference tie actions to current observations.
View repositoryI developed a cross-modal Transformer combining facial and speech representations to study emotion recognition in children with autism spectrum disorder and intellectual disabilities.
A technical analysis of how structured-output constraints can interrupt repository investigation, and an architecture that separates tool-assisted analysis from output formatting.
First Class with Distinction · 8.89 / 10.