Praateek Yadav

About Praateek Yadav

Currently contributing as a SWE Bench - Data Science at Turing, working on training, evaluating, and enhancing the performance of large language models (LLMs). Responsibilities include designing and curating machine learning-centric training data, evaluating model outputs, and developing high-signal tasks that enhance accuracy and robustness. Collaboration with cross-functional teams has been key to ensuring adherence to strict task constraints and improving real-world system behavior. With a solid foundation in Computer Science and Engineering from Dr. A.P.J. Abdul Kalam Technical University and further advanced studies in Data Science at Daksh Gurukul IIT Guwahati, expertise spans MLOps, pipelines, and model deployment. Passionate about leveraging data science to improve machine learning reasoning, data quality, and model alignment, aiming to contribute to impactful, innovative solutions in AI and analytics.

Experience

1. Swe Bench - Data Science

   Turing

   Sep 2025 — Present

   Worked on training, evaluating, and improving large language models (LLMs) with a strong focus on data quality, machine learning reasoning, and real-world system behavior. Collaborated with cross-functional research and engineering teams to design high-signal tasks that enhance model accuracy, robustness, and alignment.•Designed and curated ML-centric training data covering data science, statistics, algorithms, ML pipelines, and production-grade system scenarios•Evaluated LLM outputs for correctness, reasoning depth, hallucination risk, and adherence to strict task constraints•Built and calibrated detailed rubrics to assess model performance across dimensions such as factual accuracy, logical consistency, edge-case handling, and instruction compliance•Created prompt–response pairs and multi-step reasoning tasks aligned with supervised fine-tuning (SFT) and reinforcement learning workflows•Analyzed model failure patterns and provided structured feedback to improve generalization on complex data and ML use cases•Worked on tasks involving feature engineering, model evaluation metrics, data leakage prevention, and ML-ops considerations•Ensured high data integrity standards through iterative review, validation, and quality assurance processes•Contributed to continuous improvement of LLM behavior in analytical, coding, and decision-making contexts

Education

• Dr. A.P.J. Abdul Kalam Technical University

  B.tech, Computer science and engineering

• Daksh Gurukul IIT Guwahati

  Data Science - Credit Link Program

• SGS Inter College Banarasipur Varanasi

  Intermediate , Mathematics

• SGS Inter College Banarasipur Varanasi

  High School , Science