Rahul Kadam

Software Engineer

Brooklyn, New York, United States0 mo experience

AI EnabledAI ML Practitioner

Key Highlights

Expert in GPU efficiency and distributed training.
Proven experience in open-source contributions.
Strong background in reproducible ML tooling.

Stackforce AI infers this person is a Machine Learning Engineer with a focus on Healthcare and Open Source Development.

Contact

Skills

Core Skills

Machine LearningResearch EngineeringOpen Source DevelopmentSoftware Engineering

Other Skills

PyTorchCUDAGPU optimizationsamplingnumerical parity checksPythonworkflow standardizationpytestBLEU metricsCI/CDOpenAI CodexCLI ToolsTestingPerformance BenchmarkingDistributed Training

About

New-grad ML Systems / SWE-in-ML focused on GPU efficiency, distributed training, and reproducible ML tooling. - Merged upstream open-source PR to Opik (Comet ML): added BLEU metrics and tests/docs. - Built reproducible GPT-2 distributed training benchmark on Slurm/V100: fixed-work protocol, scaling and audit logs. - Research engineering at NYU Langone (CAI2R): PyTorch/CUDA workflows with correctness checks and evaluation metrics. Targeting New Grad / Early Career roles in ML Systems / Training Infra / SWE-in-ML Platform (NYC/US).

Experience

0 mo

Total Experience

Average Tenure

Current Experience

Nyu langone health

Research Engineer (CAI²R Lab)

Jul 2025 – Present · 10 mos · New York, New York, United States

Ported MRI reconstruction research code from MATLAB → Python/PyTorch, and added numerical parity checks to validate outputs against legacy implementations.
Implemented sampling/optimization components (e.g., BASS-style sampling + per-iteration metric logging such as k-space NRMSE / SSIM) to support systematic experiments.
Optimized training/evaluation workflow for GPU runs (V100) by standardizing configs, seeds, and run artifacts for reproducibility.

PyTorchCUDAGPU optimizationsamplingnumerical parity checksMachine Learning+1

Comet

Open Source Developer

Dec 2024 – Jan 2025 · 1 mo · New York, New York, United States · Remote

Implemented BLEU evaluation metrics (SentenceBLEU, CorpusBLEU) for Opik’s Python SDK; refactored shared logic via BaseBLEU for maintainability.
Added pytest unit tests for edge cases + error handling; changes run in the repo’s existing CI checks.
Collaborated with maintainers through PR review iterations (API conventions, validation, docs/examples).

PythonpytestBLEU metricsCI/CDOpen Source DevelopmentSoftware Engineering