-
NYU
- New York, NY
- https://deepanshumody.com
- in/deepanshu-mody-1202921b7
Pinned Loading
-
Analysis_RMP_Ratings
Analysis_RMP_Ratings PublicForked from Samarth26/Analysis_RMP_Ratings
Statistical analysis of 89,893 RateMyProfessor records — gender bias, teaching tags, and predictive models. Tested Python package + Quarto site.
Python 1
-
physiocoach
physiocoach PublicReal-time AI physical-therapy coaching: local VLM form feedback and MediaPipe rep counting & ROM. Built at the Dell × NVIDIA Hackathon 2026 (Top-8/30, NYU CDS).
Python 1
-
increasing-refusal-intervention-robustness
increasing-refusal-intervention-robustness PublicTraining-time defense that redistributes LLM refusal via mean/covariance matching + KD, raising linear-ablation attack rank from K=1 to K≥16 (Llama-3.2-1B-Instruct)
Python
-
discovery-agents
discovery-agents PublicProduction-grade multi-agent RAG workflow + an ML-infrastructure platform (PyTorch/DDP, RAG, evals, guardrails, MCP). Keyless-by-default; ships a real, reproducible Banking77 retrieval benchmark.
Python 2
-
PLM_interpretability
PLM_interpretability PublicSparse-autoencoder feature circuits in the ESM-2 protein LM: top-k SAEs + causal ablation map features to transmembrane & signal-peptide detection; interactive Cytoscape viewer
If the problem persists, check the GitHub status page or contact support.

