I'm Astha Adhikari,
an AI Engineer.
4+ years turning research into real products. From 3D perception on edge devices to LLM agents in production, I build AI that ships.
4+ years turning research into real products. From 3D perception on edge devices to LLM agents in production, I build AI that ships.
// about me
// experience
Architected production-grade LangChain agents using RAG and Vector Databases; integrated MCP tools to automate booking workflows, resulting in a 40% efficiency gain and significant reduction in API latency.
Optimised 3D LiDAR segmentation models for edge deployment on NVIDIA Jetson using TensorRT; achieved a 70% increase in inference speed while maintaining high-fidelity detection for autonomous systems.
Spearheaded research into Generative AI and Computer Vision using PyTorch, resulting in 3 peer-reviewed publications with improvements over baseline benchmarks in Image Computing.
Engineered a high-traffic Telemedicine platform and EMR system; integrated a custom NLP-based autocomplete feature using FastAPI, improving medical data entry speed by 40% while ensuring HIPAA-level data privacy.
Delivered production-ready AI applications, including Rasa-based conversational agents and Computer Vision tools for ICR and object localisation, resulting in a 60% reduction in manual data processing time.
// skills
// research
// education
// community
// contact
Whether it's a collaboration, research, or just a chat about AI — I'd love to hear from you.