The Intersection of Machine Intelligence and Scalable Systems

I am an Applied Scientist and Software Engineer dedicated to building the next generation of Agentic Systems and Foundation Models. My work lives at the intersection of deep research and production-grade engineering, focusing on creating AI that is not only intelligent but also robustly grounded in reality.

I specialize in architecting agentic pipelines and optimizing Retrieval-Augmented Generation (RAG) architectures to improve retrieval accuracy and system latency. With a background in building full-stack MLOps ecosystems, I treat infrastructure as a first-class citizen, ensuring that complex AI workloads are scalable, measurable, and reliable across enterprise environments.

Research & Scientific Impact

My research contributions span the domains of Genomics, Proteomics, Cheminformatics & Histopathology. I have co-authored several papers on specialized foundation models, including long-range genomic modeling and target-aware ligand generation. My previous experience also includes developing high-sensitivity AI modules for clinical diagnostics, successfully reducing specialist workloads through automated quantification and detection systems.

Technical Expertise

  • Agentic AI: Expertise in LangGraph, LangChain, MCP architecture, and autonomous agent workflows.
  • Foundation Models: Experience in LLaMA2-style pre-training and fine-tuning, dense self-attention, and ultra-long sequence modeling.
  • Enterprise MLOps: Proficient in Kubernetes (K8s), Helm, Terraform, and cloud-agnostic inference pipelines.
  • Multi-Modal Systems: Architecting unified backends for chat, voice-to-text, and real-time data visualization.

Featured Research

  • Chemical Language Models: Researching target-aware ligand generation and de novo molecule prediction. (OpenReview, arXiv)
  • Genomic Foundation Models: Developing long-range transformers with dense attention for variant interpretation.(OpenReview, arXiv)
  • Protein Language Models: Engineering novel protein binders through structural alignment and target-aware generation.(OpenReview, arXiv)
  • Clinical AI: Validating computer-vision models for high-sensitivity cancer detection and medical image analysis.(Cell)

Beyond the Code

Outside of building agentic systems, I am an amatuer watch collector and Gunpla enthusiast. I find that the precision required in mechanical horology and high-grade model building mirrors the discipline needed to design reliable distributed systems. I also maintain a focus on fundamental strength through weightlifting.