Kushagra Chandhok

Design and evaluate benchmark tasks for frontier large language models, focusing on reasoning, autonomous decision-making, and real-world software engineering workflows. Develop reproducible machine learning evaluation pipelines to measure model robustness, generalization, execution accuracy, and failure modes across diverse benchmark suites.

Build

Dockerized experimentation environments and Linux-based tooling to ensure deterministic evaluation, reproducibility, and scalable benchmark execution.

Machine Learning Expert - RustLabs AI

(2026-05)

• Design and evaluate benchmark tasks for frontier large language models, focusing on reasoning, autonomous decision-making, and real-world software engineering workflows.
• Develop reproducible machine learning evaluation pipelines to measure model robustness, generalization, execution accuracy, and failure modes across diverse benchmark suites.
• Build Dockerized experimentation environments and Linux-based tooling to ensure deterministic evaluation, reproducibility, and scalable benchmark execution.
• Collaborate on AI evaluation infrastructure, dataset curation, and benchmark design supporting next-generation autonomous agent research.

Software Engineering Expert - AfterQuery

(2026-03)

• Author complex SWE-Bench and Terminal-Bench style repositories for evaluating frontier AI coding agents on realistic software engineering tasks.
• Engineer reproducible Docker environments with hidden verification suites, oracle baselines, automated grading pipelines, and deterministic evaluation workflows.
• Design Linux, Git, shell scripting, filesystem, debugging, and multi-step programming challenges requiring long-horizon reasoning and autonomous execution.
• Validate benchmark quality through extensive testing, difficulty calibration, failure analysis, and compatibility across multiple state-of-the-art coding models.

Bachelor of Technology - Computer Science & Engineering (Artificial Intelligence & Machine Learning) - K.R. Mangalam University (2023 - 2027)