About
Aman Chandra H
Also written Aman Chandra · Aman C H · amanchandrah
Research Intern at IISc Bangalore · Machine learning & medical AI researcher · Freelance web developer · B.Tech (Hons) CSE, RV University
Bengaluru, Karnataka, India
Bio
Aman Chandra H (also written Aman Chandra, Aman C H, or amanchandrah) is a machine learning and medical AI researcher from Bengaluru, India. He is pursuing a B.Tech (Honors) in Computer Science and Engineering at RV University and is currently a Research Intern at Barik Labs, Centre for Neuroscience, Indian Institute of Science (IISc) Bangalore.
His research focuses on trustworthy clinical AI: calibration and evaluation of ICU mortality models under severity and prevalence shift, published in IEEE Access and with Springer (CCIS), alongside work on multilingual LLM behaviour and retrieval-augmented generation for Indian legal question answering.
He also builds software end to end: Scratcher, a YOLOv11 + FastAPI behaviour-analysis app used at IISc; NOMOS, a privacy-first RAG legal assistant; and full-stack products such as Strumrr and PlotLab. As a freelance web developer he designs and builds responsive, interactive websites with databases, making heavy use of AI tooling.
Open to
- Machine learning / AI research internships
- Applied ML and medical AI roles
- LLM, RAG and NLP engineering
- Freelance full-stack web development
Email amanchandrah@gmail.com · CV: AmanChandra_CV.pdf
Experience
Current
Research Intern — Barik Labs, Department of Neuroscience · IISc Bangalore
May 2026 — Present · Research
Research at the intersection of neuroscience and AI at one of India's premier research institutions.
Freelance Web Developer — Independent
2025 — Present · Industry
I design and build responsive, interactive websites end to end — front end, back end and database — and make heavy use of AI tooling to ship them faster.
Previously
Machine Learning Intern — Ganglia Technologies Pvt. Ltd. · Incubated at MAHE
2025 · Industry
Developed clinical severity detection pipelines; conducted literature surveys for review papers; backend dev + API integration for applied ML in healthcare.
Research Intern — Centre of IoT and Edge Computing · RV University
2024 · Research
Designed a real-time IV fluid monitoring system using non-contact capacitive sensors; integrated with IoT platform for centralised alerting (published in Springer CCIS).
Head of Technicals and Outreach — The Representatives Club
2025 · Position of Responsibility
Leads the club's technical work and its outreach.
Super-Coordinator, Outreach — Paradox 2025
2025 · Position of Responsibility
Led departmental tech efforts and volunteer management for a large-scale flagship fest.
Event Head, Generative AI — Margazhi 2025
2025 · Position of Responsibility
Organised GenAI track with 300+ participants and featured top AI artists in India.
Co-Founder — Panda Scoop
2024 — Present · Venture
Co-founded early-stage startup; Grand Finale Winner, Manthan 2024 (FKCCI); Runners Up at Idea2POC Elevate.
Education
- B.Tech (Honors), Computer Science and Engineering — RV University, Bengaluru (Pursuing)
Research & publications
- Severity-Weighted Calibration Error for Reliability Assessment Under Outcome-Severity ImbalanceIEEE Access · 2026 · Published · doi:10.1109/ACCESS.2026.3717951
- Intravenous Bag Monitoring and Alert System — A PrototypeSpringer Nature · CCIS · 2025 · Published · doi:10.1007/978-3-032-14908-4_45
- Severity-Aware Evaluation of ICU Mortality Prediction Using Risk-Weighted Error Metrics (RWEM)BMC Medical Informatics · Springer Nature · 2025 · Under Peer Review · doi:10.21203/rs.3.rs-9429272/v1
- Prevalence Shift and ICU Mortality Miscalibration Across InstitutionsICETCI 2026 · Mahindra University, Hyderabad · 2026 · Presented 18–22 Aug 2026 · doi:10.21203/rs.3.rs-10254564/v1
- Language as a Hidden Variable: Measuring Behavioral Divergence in Multilingual Large Language ModelsICETCI 2026 · Mahindra University, Hyderabad · 2026 · Presented 18–22 Aug 2026 · doi:10.21203/rs.3.rs-10218350/v1
- NOMOS: RAG Framework for Jurisdiction-Specific Indian Legal QAICETCI 2026 · Mahindra University, Hyderabad · 2025 · Presented 18–22 Aug 2026
- Clinical Risk-Weighted Evaluation of ICU Mortality Prediction Models (CRWS)ICAIHC 2026 · NIT Raipur · 2025 · Presented 18 Sep 2026
- Does Model Compression Amplify Clinical Bias? A Subgroup Fairness Audit of Quantized ICU Risk Models on MIMIC-IVIEEE INDICON 2025 · 2026 · Submitted
- Graph-Based Polypharmacy Risk Networks for Adverse Event Prediction in ICU Patients: A MIMIC-IV AnalysisIEEE DELCON, BITS Pilani · 2026 · Submitted
Projects
- Scratcher — A full-stack ML app that turns raw mouse-session video into quantified scratching data — one click from detect to publication-ready figures. (FastAPI, YOLOv11, PyTorch, Full-Stack)
- NOMOS — A privacy-first RAG legal-QA system for Indian statutory law — cited, jurisdiction-specific answers with fully local inference. (React, Flask, LangChain, Llama 3.2)
- Panda Scoop — An early-stage startup focused on innovative solutions. (Business Plan, Market Research, Startup)
- IV Bag Monitoring System — IoT-based remote monitoring system for intravenous fluids. (IoT, Healthcare, Research)
- Eventify — A comprehensive event management platform — create, host, ticket. (React, MongoDB, Node.js)
- Strumrr — A full-stack guitar practice platform — daily logging, streaks, heatmaps, and instructor video feedback with browser-side compression. (React Router 7, Hono, Turso, Cloudinary, Tailwind CSS)
- Genreize — A music genre classification platform using HuBERT transformer model. (PyTorch, TorchAudio, Deep Learning)
- PlotLab — A no-code, offline scientific figure generator — spreadsheet in, journal-ready plots out. (Python, FastAPI, matplotlib)
- Outreach Portal — An internal AI-powered guest handling & request system for a large-scale fest. (React, MongoDB, Node.js)
Skills
- Languages & Core
- Python, C, C++, Java, JavaScript, SQL
- ML / DL
- PyTorch, TensorFlow, Scikit-learn, XGBoost, LightGBM, Pandas, NumPy, LSTM
- Medical AI & Calibration
- SOFA / APACHE, ECE & Brier Score, SHAP, Isotonic Regression
- LLM / RAG
- LangChain, ChromaDB, Ollama, Hugging Face, Sentence-Transformers
- Web & Backend
- React.js, Node.js, Express.js, HTML5, CSS3
- Cloud & Platforms
- Firebase, Google Cloud, Apps Script
- Tools
- Git, GitHub, Docker, LaTeX, MySQL, MongoDB, Arduino
Profiles
Frequently asked
Who is Aman Chandra H?
Aman Chandra H is a machine learning and medical AI researcher from Bengaluru, India. He is a Research Intern at Barik Labs, Centre for Neuroscience, IISc Bangalore, and is pursuing a B.Tech (Honors) in Computer Science at RV University. His work has been published in IEEE Access and with Springer.
What research has Aman Chandra H published?
His papers include 'Severity-Weighted Calibration Error for Reliability Assessment Under Outcome-Severity Imbalance' (IEEE Access, 2026), 'Intravenous Bag Monitoring and Alert System — A Prototype' (Springer CCIS, 2026), and work on risk-weighted evaluation of ICU mortality prediction (RWEM, CRWS), prevalence shift and ICU miscalibration, multilingual LLM behavioural divergence ('Language as a Hidden Variable'), and NOMOS, a RAG framework for Indian legal question answering.
What does Aman Chandra H work on?
Clinical and medical AI (model calibration, evaluation, ICU mortality prediction), computer vision for neuroscience (Scratcher, at IISc), LLMs and retrieval-augmented generation, and full-stack web development as a freelancer.
Where does Aman Chandra H work and study?
He studies at RV University, Bengaluru (B.Tech CSE) and is a Research Intern at Barik Labs, IISc Bangalore. He previously interned at Ganglia Technologies (incubated at MAHE) and at the Centre of IoT and Edge Computing, RV University.
Is Aman Chandra H open to internships or freelance work?
Yes — he is open to machine learning and AI research internships, applied ML and medical AI roles, LLM / RAG engineering, and freelance full-stack web development. The quickest way to reach him is email: amanchandrah@gmail.com.
What is Panda Scoop?
Panda Scoop is an early-stage startup Aman Chandra H co-founded. It placed 4th at Manthan 2024, the FKCCI Karnataka business plan competition with over 300 teams, and was Runners-Up at Idea2POC Elevate.