Photo of Siddhant

Siddhant Saxena

Contact: (+91) 7024184657
Email:
Portfolio: My Portfolio
LinkedIn: linkedin.com/in/siddhant-saxena
GitHub: github.com/mr-siddy

I am a pre-doctoral researcher specializing in Geometric Deep Learning, Neural Algorithmic Reasoning, Reinforcement Learning, Large Language Models, Computer Vision, and Combinatorial Optimizations.

Education

Pre-Doctoral Fellowship

Institute: Indian Institute of Technology Delhi (IITD)

Duration: Aug 2023 – June 2024

Specialization: Geometric Deep Learning

Bachelor of Technology (B.Tech)

Institute: Manipal Academy of Higher Education

Duration: Aug 2019 – Apr 2023

GPA: 9.26/10

Experience

Applied Scientist, Snapdeal

Duration: June 2024 – Present

  • Developed a temporally evolving multi-modal graph recommendation system.
  • Implemented Graph-RAG architecture for tabular data and Return-to-Origin problem.

Pre-Doctoral Researcher, LCS2-IITD cw/ META AI

Duration: Aug 2023 – June 2024

  • Researched learning on higher-dimensional Riemannian manifolds.
  • Continual learning on graphs using teacher agents.

Lead Researcher, LCS2-IITD cw/ Meesho

Duration: Jan 2023 – Aug 2023

  • Designed novel message propagation techniques for large-scale hypergraph topology.

Deep Learning Research Fellow, BMCHRC

Duration: Nov 2021 – Apr 2022

  • Conducted research on Geometric Manifold Generation for cancer studies.

Algorithmic Design Intern, Aspire Research Park

Duration: Dec 2019 – Jan 2020

  • Developed algorithms to improve code readability and computational efficiency.

Software Developer Intern, CueMath

Duration: May 2020 – Jun 2020

  • Automated report generation processes using feedback analysis.
Skills

Languages: Python, C++, C, Java, Julia, JavaScript

Frameworks: PyTorch, TensorFlow, FastAI, Django, Flask

Other Skills: Competitive Programming, Machine Learning, Graph Networks

Publications
  • Task-aware Latent Graph Rewiring: Addressing OSQ-OSM Dilemma
  • ImmuneGAN: A GAN-based architecture for network intrusion detection
  • Biases and De-Biasing Strategies in Graph RecSys
  • Efficient Garbage Monitoring System using GRUBin
Projects
  • FAITH: Reducing hallucinations in dialogues with Knowledge Graphs
  • DreamSpace: User-preferential floor planning via LLMs
  • PersonaKG-RAG: Personalized Recommendation Systems using Graphs
  • MRecSys: Multimodal Graph Recommendation System
Awards
  • President's Gold Medal for B.Tech Thesis
  • INSPIRE National Award - AIR 7
  • Winner - Learn IT Codeathon