Anirudh Iyengar K N

Hello and welcome!

I’m Anirudh Iyengar Kaniyar Narayana Iyengar, currently working as an AI Capacity Engineer at eBay. I build and optimize large model systems for performance, reliability, and cost.

I completed my Master’s in Robotics and Autonomous Systems in Artificial Intelligence at Arizona State University, where I worked on research publications and projects at CORAL Lab, ASU involving multimodal learning and vision-language reasoning.

My interests include machine learning, large language models, computer vision, and data-driven systems. I enjoy solving problems through clean design, careful experimentation, and efficient engineering.

Outside of work, I enjoy connecting with people from diverse backgrounds during my travels, learning from their stories and experiences. I also find joy in painting, cooking, and baking, which allow me to express creativity and mindfulness in simple ways.

For a glimpse into my personal side, visit A peek into my life experiences..

Access my structured notes and guides here:
Scrape Codebook →

Selected publications

EMNLP 2025
TABARD: A Novel Benchmark for Tabular Anomaly Analysis, Reasoning and Detection

Manan Roy Choudhury, Anirudh Iyengar Kaniyar Narayana Iyengar, Shikhhar Siingh, and 2 more authors

In EMNLP, ACL ARR (May Cycle), 2025

Abs Bib PDF Code Website

We study the capabilities of large language models (LLMs) in detecting fine-grained anomalies in tabular data. Specifically, we examine: (1) how well LLMs can identify diverse anomaly types—including factual, logical, temporal, and value-based errors; (2) the impact of prompt design and prompting strategies; and (3) the effect of table structure and anomaly type on detection accuracy. To this end, we introduce TABARD, a new benchmark constructed by perturbing tables from WikiTQ, FeTaQA, Spider, and BEAVER. The dataset spans multiple domains and eight anomaly categories, including paired clean and corrupted tables. We evaluate LLMs using direct, indirect, and Chain-of-Thought (CoT) prompting. Our results reveal notable limitations in standard prompting, especially for complex reasoning tasks and longer tables. To overcome these issues, we propose a unified framework combining multi-step prompting, self-verification, and constraint-based rule execution.
@inproceedings{tabard2025, title = {TABARD: A Novel Benchmark for Tabular Anomaly Analysis, Reasoning and Detection}, author = {Choudhury, Manan Roy and Kaniyar Narayana Iyengar, Anirudh Iyengar and Siingh, Shikhhar and Sugeeth, Raghavendra and Gupta, Vivek}, booktitle = {EMNLP, ACL ARR (May Cycle)}, year = {2025}, url = {https://openreview.net/forum?id=8Y1G4XD3vR}, data = {https://github.com/TABARD-emnlp-2025/TABARD-dataset}, }

News

Aug 01, 2025	🎉 We’ve submitted two papers: one to EMNLP 2025 and one to AACL 2026.
Aug 01, 2025	🎉 TABARD our EMNLP 2025 submission has been accepted. 📄 Proud to be a co-authored! 🙌
Apr 01, 2025	Joined CORAL Research Lab
May 31, 2024	I am excited to share that I have started a new role as a Machine Learning Intern at axyo (Prev - Synapse Lab)! ✨😊
Dec 31, 2023	I’m thrilled to announce that I have begun a new role as a Graduate Student Assistant (GSA) at Arizona State University (ASU).

Hello!

Anirudh Iyengar K N

Hello and welcome!

Selected publications

News

Skills

Python

PyTorch

C++

SQL

MatLab

R

Tableau

Pandas

Numpy

Docker

Git

OpenCV

HuggingFaces

Open AI

Jupyter Notebook

VS code

Scikit learn

OpenMM

Anaconda

Quick Sight

AWS

Colab

Kubernetes

Google Cloud

MLflow

Jenkins

Kubeflow

Data
Version Control

GitHub

Redis

Grafana

Prometheus

Hello!

Hello and welcome!

Selected publications

News

Skills

Python

PyTorch

C++

SQL

MatLab

R

Tableau

Pandas

Numpy

Docker

Git

OpenCV

HuggingFaces

Open AI

Jupyter Notebook

VS code

Scikit learn

OpenMM

Anaconda

Quick Sight

AWS

Colab

Kubernetes

Google Cloud

MLflow

Jenkins

Kubeflow

DataVersion Control

GitHub

Redis

Grafana

Prometheus

Data
Version Control