Systems
Flagship case studies first. Archive entries extend the record of systems shipped without implying a four-project career.
Production multimodal generation platform giving users unified access to a large ecosystem of image, video, and audio models—with side-by-side comparison from a single prompt.
LangGraph-orchestrated clinical platform with specialized agents, Agentic RAG, tenant isolation, human-in-the-loop gates, and evaluation via LangSmith and RAGAS.
LLM-driven system that transforms natural-language campaign requirements into structured geographic intent, retrieves inventory, and ranks placements across 290+ UAE categories in under five seconds.
Unified heterogeneous PDF, image, CSV, and waveform inputs into a canonical clinical schema with configurable field mapping—eliminating recurring manual analyst work and enabling analytics/ML pipelines.
Additional systems across retail, healthcare, recruitment, legal data, and applied NLP.
No-code conversational AI platform enabling organizations to configure and launch custom domain assistants—reducing dependence on one-off chatbot development.
Voice and conversational workflows integrating Llama, NLP pipelines, Twilio, APIs, and business actions for support, booking, information retrieval, and automation.
NLP applications using GPT-3, Llama, BERT, and Falcon—applying transfer learning and fine-tuning to domain-specific conversational and language-processing use cases.
Machine-learning and time-series forecasting system for retail sales prediction, promotional impact analysis, and stockout risk—built on FreshRetailNet-50K with a FastAPI surface and Supabase/PostgreSQL storage.
AI-powered bioinformatics platform for genomic variant interpretation, drug-response prediction, biomarker discovery, molecule generation, and NLP literature mining across gene–disease and drug–target relationships.
AI-powered recruitment platform that uses voice agents to run initial screening calls—capturing candidate intent, structuring responses, and accelerating early-funnel evaluation before human interviewers engage.
End-to-end pipeline for normalizing, discovering, and consolidating heterogeneous legal datasets into unified Golden Schemas—using agentic LLM workflows for semantic understanding and conflict resolution across disparate sources.
