Skip to content
View oshindutta's full-sized avatar
🎯
Focusing
🎯
Focusing
0 followers · 3 following

Achievements

Achievement: Arctic Code Vault Contributor

Achievements

Achievement: Arctic Code Vault Contributor

Block or report oshindutta

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
oshindutta/README.md

👋 Hi, I'm Oshin Dutta !

Ph.D. Scholar in AI at IIT Delhi • Researcher in Efficient AI • Enthusiast in LLMs, NAS, and Model Compression

🌐 Website • 📫 Email

🔬 Current Role

Ph.D. Scholar, IIT Delhi
Working on Efficient AI & Accelerating LLMs
Collaborated with Samsung Research and Cadence India

  • 🧠 VTrans – 10× speed-up for LLM fine-tuning + 50% compression
  • 🚀 TVA-prune – 60% GPU inference speed-up for LLaMA/Mistral
  • 🤖 DCA-NAS – 5× faster hardware-aware NAS on distributed GPUs

🧑‍🔬 Past Experience

Student Researcher, IIT Dhanbad

  • Worked on tempo and rhythm extraction in polyphonic music using ML
  • Published findings in IEEE conference

Intern, Aerospace Dept., IISc Bangalore

  • Designed algorithms for fuel-efficient lunar landings
  • Benchmarked efficiency on TMS320C6748 DSP

🎓 Education

  • 🎓 Ph.D. in Efficient AI, IIT Delhi (2019–2025)
  • 🎓 M.Tech., IIT Dhanbad – Electronics & Communication (2016–2018)
  • 🎓 B.E., VTU – Electronics & Communication (2011–2015)

🛠️ Skills

  • Languages: Python, C, Java, MATLAB
  • Frameworks: PyTorch, TensorFlow, OpenCV
  • AI: CNNs, RNNs, GANs, LLMs, ViTs, Multimodal, NAS
  • Research Interests: Efficient AI, Model Compression, Pruning, Quantization, NAS
  • Other: LoRA, Few-shot Learning, Post-Training Quantization, Deployment

📫 Connect with Me

LinkedIn • Twitter • GitHub

Pinned Loading

  1. TVAprune Public

    [ICML 2024 Es-FoMo] - Efficient LLM Pruning with Global Token-Dependency Awareness and Hardware-Adapted Inference

    Python 4 3

  2. DCA-NAS Public

    [PReMI 2023]- Device-Constraint - Aware Neural Architecture Search Method. It incorporates methods to constrain architecture search given device constraints and to fasten the search.

  3. CoFiPruning_RemovedErrors Public

    Forked from princeton-nlp/CoFiPruning

    ACL 2022: Structured Pruning Learns Compact and Accurate Models https://arxiv.org/abs/2204.00408

    Python 1

  4. Compression-Related-Papers Public

    Lists papers read related to model compression of transformers, CNNs, RNNs and Neural Architecture search (NAS). Includes papers on Variational Information Bottleneck.

    1

  5. tempo-estimation Public

    Matlab code to estimate tempo of various genres of music

    MATLAB