Headline

Finding Answers.

About

Final-year Computer Science engineer building production machine-learning systems end to end — local-first inference, retrieval pipelines, real-time computer vision and applied NLP research. Published at IEEE CONECCT 2026 on explainable trajectory prediction, with a secondary track in robotics and embedded systems. Recent work: • ARI Sync — real-time, local-first AI coaching engine: two 7B LLMs, Whisper and MediaPipe running entirely on Apple Silicon in under 10 GB with no cloud dependency; 30 FPS pose tracking, ~1.9 s from error onset to audible voice correction. • Semantic Movement Index (SMI) / LexiVerse — legal-tech NLP platform quantifying how legal concepts shift meaning across time and jurisdictions, validated across 265,000+ documents from the US, India and Europe. • Textigrity — eight-model content-integrity ensemble (DeBERTa, RoBERTa, XLM-R, Longformer) with a closed-loop adversarial humanizer exposing detector robustness gaps. • VidyaTid (vidyatid.in) — offline-first RAG learning companion for JEE/NEET, indexed across 49,652 questions from 20 years of past papers. • AI Raitha Mitra — crop-disease identification from leaf photos, 88% accuracy across 20 crop types, with a 3-language assistant. Also: organizer of the MLX machine-learning hackathon (~260 participants, 50+ submissions), robotics builder (ROS, Arduino, embedded systems), and certified in IBM Containers & Kubernetes and Microsoft Azure Quantum. Open to AI/ML internships and research roles — remote or on-site, available immediately. adithyanmurthy@gmail.com · github.com/Adithyanmurthy · devpost.com/adithyanmurthy

Location

Greater Bengaluru Area

AI stack

anthropic apis

claude code

claude

codex

gemini

grok

Work experience

Research Assistant at RV University, Jan 2026 - May 2026 · 5 mos, ARI Sync — real-time, local-first AI coaching engine (Core Engineering · Guide: Prof. Monesh N). • Converted instructional video into live form correction, running the entire ML stack locally on Apple Silicon — two 7B LLMs, Whisper and MediaPipe in under 10 GB with no cloud dependency. • Engineered the real-time feedback pipeline: 30 FPS pose tracking, ~2.5 ms median form-check latency, ~1.9 s from error onset to audible correction. • Architected async ingestion on FastAPI, Celery, PostgreSQL and Redis; React front-end running MediaPipe in WebAssembly for client-side pose detection.

Research Fellow at RV University, Feb 2024 - May 2024 · 4 mos, Semester-long quantum computing research fellowship (Quantum Sciences) under mentor Bharath Manchikodi. • Earned the Microsoft Azure Quantum certification, working hands-on with Azure Quantum services and Q# programming. • Studied quantum mechanics, gates and algorithms via the Linux Foundation course; authored a 50-page report on applications in cryptography and optimization.

Education

Bachelor of Technology - BTech (Hons.), Computer Science & Engineering, Rv university

Public links

github: https://github.com/Adithyanmurthy

adithyanmurthy.com: https://adithyanmurthy.com/