Skip to content
View kjgpta's full-sized avatar
2 followers · 0 following

Achievements

Achievement: YOLOAchievement: Pull Shark

Achievements

Achievement: YOLOAchievement: Pull Shark

Block or report kjgpta

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
kjgpta/README.md

AI Banner

👋 Hi, I'm Kshitij Gupta!

ML Engineer | LLM & NLP Enthusiast | AI Researcher

LinkedIn Google Scholar

🏆 About Me

🔹 Role: Machine Learning Engineer @ Chubb Business Services India

🔹 Education: B.E. in Electronics & Electrical Engineering, BITS Pilani

🔹 Research: Deep Learning, NLP, Code-Switching NLP (ACL, AACL-IJCNLP, IALP, AISC, ACIIDS)

🔹 Interests: LLM reasoning, Retrieval-Augmented Generation (RAG), Efficient Inference

💡 Current Focus

Focus
  • 🚀 Speculative Decoding for ultra-low-latency LLM outputs
  • ⚙️ Hierarchical Loss Functions to speed up convergence
  • 🌐 Distributed inference and system-level optimizations

🚀 Recent Projects

Project Description Status
🕵️ WhoDunIt Benchmark for deduction & reasoning in LLMs. ✅ Live
⚙️ EfficientLLM Speculative decoding & hierarchical loss modules. 🚧 WIP

🛠️ Tech Stack

Languages & Frameworks

Python | Java | C++ | JavaScript PyTorch | TensorFlow | Keras | Transformers DeepSpeed | vLLM | OpenAI API | LangChain

Infrastructure & MLOps

Azure ML Studio | Databricks | GCP Vertex AI | AWS SageMaker MLflow | Kubeflow | Docker | Kubernetes | AKS

Data & DevOps

PostgreSQL | MongoDB | Apache Spark | Kafka | Redis Git, GitHub Actions, Jenkins | Linux | Bash | Jupyter

📚 Publications

  • WhoDunIt: Culprit detection benchmark (ACL ARR, 2024)
  • Code-Switching NLP (IALP, 2023)
  • Statistical Text Augmentation (ACIIDS, 2023)
  • MALM: Zero-Shot MT (AACL-IJCNLP, 2022)
  • Essay Scoring with Transformers (AISC, 2023)

🌐 Let's Connect!

LinkedIn

Email

“AI is our most powerful tool—let’s use it wisely!” 🚀

Pinned Loading

  1. kjgpta.github.io kjgpta.github.io Public

    https://kjgpta.github.io/

    CSS 1

  2. NSC-Code-Switch-Analysis NSC-Code-Switch-Analysis Public

    Code-switching analysis using linguistic techniques like part-of-speech

    Jupyter Notebook

  3. Code-Switch-Language-Modeling-for-English-and-Malay Code-Switch-Language-Modeling-for-English-and-Malay Public

    Code-Switched Data generation based on Part-of-speech and Language Modeling of the generated text.

    Jupyter Notebook 1

  4. Data-Augmentation-for-Automated-Essay-Scoring-using-Transformer-Models Data-Augmentation-for-Automated-Essay-Scoring-using-Transformer-Models Public

    A novel data augmentation approach for Automated Essay Scoring and modeling it using Tranformers model

    Jupyter Notebook

  5. Unity-Projects Unity-Projects Public

    In this Repository, you will find projects such as the roll-a-ball games in Unity 3D.

    C#

  6. Token-Bucket-Algorithm Token-Bucket-Algorithm Public

    Token Bucket Algorithm with GUI interface

    Java