sid0nair@research— incoming MS Privacy Eng. @ CMU
~/ cv

Curriculum Vitae

00

Summary

B.Tech in Mechanical Engineering with a Computer Science minor at IIT Delhi, specializing in privacy-preserving machine learning and federated optimization. Research in differential privacy, federated learning and secure AI systems across IIT Delhi, IIT Hyderabad and ISI Kolkata, with work accepted at TMLR and under review at IEEE T-PAMI. Core expertise in second-order federated optimization with DP guarantees, privacy-accounting mechanisms, and privacy-preserving system deployment. Incoming MS in Privacy Engineering at Carnegie Mellon.

01

Education

Carnegie Mellon University MS in Privacy Engineering · School of Computer Science
2026 — incoming
Indian Institute of Technology Delhi B.Tech in Mechanical Engineering + CS Minor · CGPA 7.92 / 10.0
Oct 2022 — 2026
Modern Vidya Niketan, Faridabad Class XII — CBSE · 96.8%
Mar 2022
Modern Vidya Niketan, Faridabad Class X — CBSE · 97.2%
Mar 2020
02

Scholastic Achievements

03

Publications & Patents

detailed →
under review

Accelerated Training of Federated Learning via Second-Order MethodsIEEE T-PAMI · Apr 2025

arXiv:2505.23588
under review

FedDAF: Federated Domain Adaptation using Model Functional DistanceSpringer Machine Learning · Aug 2025

arXiv:2509.11819
published

Towards Support-Free Printing in Extrusion-Based Additive ManufacturingASME IMECE 2025 · Mar 2025

IMECE-160285
04

Research Experience

Differentially Private 2nd-Order Optimization in FL — ISI Kolkata
Jun 2025 — Present
Prof. Tanmay Sen · Statistical Quality Control & OR Unit
  • Designing 2nd-order FL optimizers with momentum-corrected Newton steps for faster convergence.
  • Implementing a federated privacy accountant with Rényi DP and subsampling amplification, ensuring (ε, δ)-DP.
  • Building a dimension-aware gradient-clipping framework for ResNet, stabilizing updates under noisy Hessians.
  • Achieved 8× noise reduction via subsampling — 70% accuracy, outperforming first-order baselines.
Accelerated Training of FL via 2nd-Order Methods — Indian Institute of Technology (IIT), Hyderabad
Oct 2024 — Apr 2025
Prof. C. Mohan · Dept. of CSE
  • Benchmarked convergence of Newton-CG, L-BFGS and quasi-Newton methods on heterogeneous federated clients.
  • Implemented Hessian-free and low-rank approximation schemes, reducing communication overhead by 40%+.
  • Tested diagonal and Nyström-based second-order updates for scalability in large-scale distributed optimization.
  • Developed preconditioning strategies for distributed second-order optimization with non-IID data heterogeneity.
05

Work Experience

Healthcare Design & Innovation Intern — Philips India Ltd
May 2025 — Jul 2025
Dr. Amar Banerjee · R&I Management, HIC Pune
  • Built an AI-driven CAD pipeline converting natural-language prompts to structured JSON for parametric modeling.
  • Integrated SQL-backed ChromaDB retrieval with the Fusion API to align JSON parameters for precise code mapping.
  • Deployed Qwen 2.5 Coder (7B) on Ollama, translating JSON into Autodesk Python APIs for automated CAD tasks.
  • Validated the CAD–SQL workflow across tasks, cutting manual drafting by 60% and boosting iteration speed.
SSUP Research Intern — SONY Corp.
Apr 2024 — Oct 2024
Prof. Sagar Sarkar & Prof. Amber Srivastava · Mech. Eng., IIT Delhi
  • Designed a native slicing & G-code framework in Blender for India's first 5-axis FDM 3D printer.
  • Integrated Python APIs (PyDy, BlendPy) with Rhino3D workflows for precise multi-axis toolpath planning.
  • Optimized slicing routines to minimize redundant tool movements, improving deposition efficiency & accuracy.
  • Implemented CNN and ViT based anomaly detection for extrusion, enabling adaptive feedback & early fault alerts.
06

Technical Skills

Programming
PythonC / C++SQLMATLAB
AI / ML
VAE-GANRényi DPBiLSTML-BFGSNewton-CGDiffusion (Wavelet, DreamFusion)
Neuroimaging
fMRIBOLDGLMFSL / SPM
Tools
PyTorchTensorFlow PrivacyOpacusChromaDBLangChainOllamaGitCuPyXGBoost
CAD / Sim
Fusion 360AnsysBlenderRhino3D
07

Positions of Responsibility

08

References

Prof. Rahul Garg
Professor, Dept. of Computer Science & Engineering, IIT Delhi
Prof. Chalavadi Krishna Mohan
Professor, Dept. of Computer Science & Engineering, IIT Hyderabad
Prof. Tanmay Sen
Assistant Professor, SQC & OR Unit, ISI Kolkata
Prof. Sagar Sarkar
Assistant Professor, Dept. of Mechanical Engineering, IIT Delhi