Yuvraj Singh

Yuvraj Singh

NLP | Interested in RL and Distributed Systems | Exploring the intersection of Research and Engineering

Hi there, I am Yuvraj!

  • Build and fine-tune LLMs; re-implement classic AI/ML papers from scratch
  • Focus areas: Multimodal LLMs (vision, language and audio) and RL for pre/post training paradigms
  • Looking for: Research Internship (academia / industrial) positions in my areas of interest. Open to AI Engineer roles for development of pipelines, RAGs, Fine Tuning and other AI/ML related projects
  • Community: Mentor newcomers and collaborate on open-source projects

Professional Experience

  • TurboML— AI Engineer · May 2025 – June 2025
    • Worked on development of various tools for LLM, enhancing tool-calling capabilities.
    • Developed pipeline for smooth integration of tools developed for YouTube QnA, summarization of videos and fact check features with the platform.
  • University of Maryland — Research Intern · Dec 2024 – Feb 2025
    • UI/UX→Code dataset creation for VLM fine‑tuning
    • Scraped 100+ static sites; curated 200+ layout/commit records
  • IISER Kolkata — Summer Research Intern · May 2024 – May 2025
    • Built 40k human‑verified stance‑analysis dataset (sports controversies)
    • Labels via Llama‑3.1/3.2, Mistral‑7B, Qwen‑2.5 with human verification
    • Fine‑tuned reasoning and non‑reasoning LLMs; ~30% improvement
  • AIISC — Research Intern · Mar 2024 – Jul 2024
    • Worked with Prof. Amitava Das on hallucination prevention
    • Web‑scraped news/posts; generated synthetic data with open‑source LLMs
    • Entity tagging using GLiNER and related models
  • Clinical AI Assistance — Research Intern · Dec 2023 – Mar 2024
    • Dataset creation and model evaluation with HuggingFace‑hosted LLMs
    • Data gathering, cleaning, and efficiency analysis

Projects

  • SmolCluster
    Distributed Deep Learning Library for Heterogeneous Hardware
    2026-01-01 00:00:00 -0800
    Research Library · Personal Project · Online
    • Educational Library for training/inference of neural networks across heterogenous compute like Mac minis, Raspberry Pi, and GPUs, written using only socket library in Python.
    • Supports FSDP, Classic Data Parallelism, Elastic DP, Synchronous Parameter Server, and Model Parallelism.
  • NeatRL
    Deep Reinforcement Learning Algorithms Library
    2025-02-01 00:00:00 -0800
    Research Library · Personal Project · Online
    • Comprehensive implementations of deep RL algorithms including DQN, A2C, PPO, DDPG, TD3, and SAC.
    • Features one-file implementations, experiment tracking with W&B, automatic video recording, and support for Gymnasium environments.
    • Main NeatRL library provides high-quality training utilities with focus on simplicity and performance.
  • NeatRL Playground
    AI Games Showcase powered by Reinforcement Learning
    2025-01-01 00:00:00 -0800
    Interactive Platform · Personal Project · Online
    • Beautiful, interactive website showcasing AI-powered games with reinforcement learning agents.
    • Features Pong AI with Deep Q-Learning, real-time WebSocket communication, and smooth animations.
    • Deployed on Vercel (frontend) and Render (game server) with production-ready health checks and headless mode.
  • Paper Replications
    ML/DL Research Paper Implementations
    2025-01-01 00:00:00 -0800
    Research Repository · Personal Project · Online
    • A comprehensive collection of code implementations replicating results from influential machine learning and deep learning research papers.
    • Features 30+ models including Transformers, GANs, Vision models, and RLHF techniques.
  • FarmGenie (GeoHack 2024)
    Empowering farmers with real-time insights and expert guidance via AI-driven space
    2024-07-19 00:00:00 -0700
    Hackathon · IIIT-BH · Bhubaneswar
    • Our platform utilizes LLMs and a Mixture of Expert (MoE) approaches to provide precise guidance on soil management, plant disease identification, and irrigation techniques.
    • Built as a scalable web application with a Next.
    • js frontend and backend, and supported by a Redis queue and multiple worker nodes, FarmGenie ensures robust performance.
  • Insight-Ed (HackNITR 5.0)
    EdTech Platform for Student and Teacher
    2024-03-03 00:00:00 -0800
    Hackathon · NIT · Rourkela
    • Imagine an online classroom where teachers instantly know when and why students lose focus.
    • Our AI-powered solution bridges the knowledge gap by detecting student emotions and attentiveness, highlighting problem areas, makes the teacher aware of each student's progress.
    • With features like reverse video search, dynamic questionnaires, and advanced Q&A bots, we transform the learning experience, making it more interactive and insightful.
  • PlogPayouts
    AI-driven Plogging System
    2023-12-01 00:00:00 -0800
    Google's Solution Challenge '24 · · Bhubaneswar
    • Transform your daily jog into a mission for a cleaner world with PlogPayouts.
    • Our innovative website + app rewards you for collecting litter, promoting fitness and environmental cleanliness.
    • Utilizing AI for trash categorization and optimized routes, and fostering community through shared stories, PlogPayouts turns every step into a step towards a greener, more inclusive society.
  • MoviesMania (Geek-o-thon)
    A Reverse Search based Movies Recommendation System
    2023-10-03 00:00:00 -0700
    Hackathon · IIIT · Bhubaneswar
    • Step into the future of entertainment discovery with MoviesMania.
    • The rpoduct aims to simplify your search for the perfect movie or web series.
    • Using various AI/ML techniques and elements, we analyze uploaded video clips to predict movie titles and recommend similar content with an impressive accuracy.
  • Movies Review System
    Spoiler-Free Sentiment-Analysis based Movies Review System)
    2023-09-01 00:00:00 -0700
    POC · IIIT · Bhubaneswar
    • Introducing the Movie Review System, where AI meets movie magic to revolutionize how viewers experience films.
    • This project goal is to provide an interface for spoiler-free reviews and sentiment analysis, enhancing the viewing journey.
    • With advanced models like Voting Classifier and Bi-LSTMs powered by Keras and TensorFlow, we achieve impressive metrics—a 91% accuracy, 91% precision, and 90% recall that understands and enhances the users' with the respective movies' plot from a bird's eye view.

Education

  • International Institute of Information Technology, Bhubaneswar · 2023–2027
    • BTech, Computer Science Engineering
  • Delhi Public School · 2022–2023
    • CBSE Grade 12 — 91%
  • Amity International School · 2021–2022
    • CBSE Grade 10 — 96%

Achievements

  • GeoHack '24 Finale — 2nd place Project: FarmGenie · IEEE GRSS Kolkata and SAADRI · 2024
  • YESIST12 '24 (Special Track) — Finalist Project: PlogPayouts · Led the project · 2024
  • Geek‑o‑thon (D3 @ IIIT‑BH) — Winner
    Project: MovieMania · Inter‑college hackathon (AI/ML) · 2023