Research Engineer, Fast-Track Training & Graph Learning

Monee; Singapore; On-site; SGD 120,000 - 180,000; Full time

Job summary

Monee is seeking a Research Engineer to bridge model research and engineering, building reliable experimentation and training foundations for behavioral sequence pretraining and graph learning. You will tackle bottlenecks in data loading, sampling, computation, memory, and evaluation, while improving turnaround, resource use, and observability for researchers.

You will contribute to reproducible workflows, performance baselines, and efficient sequence training across multi-GPU settings,

Qualifications

• Bachelor's degree or above in Computer Science, Software Engineering, AI, or a related field with strong software foundations and demonstrable research engineering, ML engineering, or systems projects.
• Proficiency in Python and PyTorch with understanding of training loops, autograd, batches, optimizers, checkpoints, and inference, and the ability to debug model code.
• In-depth experience in training systems, data pipelines, performance diagnosis, distributed systems, or experimentation infrastructure with clear bottlenecks, trade-offs, and improvements.
• Foundational performance analysis skills across GPUs, CPUs, memory, and I/O, using profilers and runtime metrics to identify issues.
• Attention to model quality, experimental comparability, numerical correctness, data versions, and reproducibility for engineering changes.
• Strong practices in code organization, testing, logging, recovery, and documentation, with ability to work with researchers on open-ended problems.
• Ability to read technical material, write engineering docs, and collaborate across teams.

Responsibilities

• Build reproducible experimentation workflows for training, checkpointing, recovery, evaluation, and result recording; track code, data, graph snapshots, sampling, and randomness.
• Profile data loading, sampling, transfer, forward/backward passes, synchronization, and output writing; establish benchmarks for throughput, memory, and turnaround to target quality.
• Improve sequence training efficiency through data sharding, prefetching, bucketing, padding, batching, mixed precision, and checkpointing; evaluate bottlenecks across backends.
• Advance graph learning engineering by optimizing storage, neighbor sampling, feature fetching, caching, CPU–GPU transfer, and training pipelines with fixed snapshots for repeatable evals.
• Develop evaluation and diagnostic tools linking training, embedding generation, downstream integration, and reporting; expose losses, gradients, anomalies, and statistics.
• Estimate capacity and compute budgets, compare single-run vs parallel experiments, and improve recovery and execution efficiency.
• Support reliable integration with application and platform teams; maintain benchmarks, tools, runbooks, and reusable code.

Skills

Python

PyTorch

ML engineering

System design

Education

Bachelor's degree in CS

Tools

DGL

PyG

CUDA

Job description

Monee is seeking a Research Engineer to bridge model research and engineering, building reliable experimentation and training foundations for behavioral sequence pretraining and graph learning. You will tackle bottlenecks in data loading, sampling, computation, memory, and evaluation, while improving turnaround, resource use, and observability for researchers.

You will contribute to reproducible workflows, performance baselines, and efficient sequence training across multi-GPU settings,