Senior / Staff SLM & VLM Engineer - Post-Training, Tool Calling & Agents

JABIL CIRCUIT (SINGAPORE) PTE. LTD.

Pasir Ris, Singapore
Not specified
Python and c++ software engineering skills
Pytorch ml training pipeline development
Model compression and quantization techniques
JABIL CIRCUIT (SINGAPORE) PTE. LTD. is seeking a Senior/Staff SLM & VLM Engineer to lead research and development efforts in Small Language Models and Vision-Language Models, focusing on low-latency and cost-efficient solutions. The role requires strong software engineering skills and experience in model optimization, with responsibilities encompassing training, compression, tool calling systems, and data pipeline automation

Job Summary

  • The role involves leading the R&D of SLMs and VLMs specifically optimized for edge and low-latency production scenarios.
  • Candidates will architect production-grade tool calling frameworks including cataloging, routing, validation, and observability.
  • The position requires strong bilingual communication skills in both Chinese (Mandarin) and English to liaise with counterparts in China.

Matching Summary

Match Score: 85

JABIL CIRCUIT (SINGAPORE) PTE. LTD. is seeking a Senior/Staff SLM & VLM Engineer to lead research and development efforts in Small Language Models and Vision-Language Models, focusing on low-latency and cost-efficient solutions. The role requires strong software engineering skills and experience in model optimization, with responsibilities encompassing training, compression, tool calling systems, and data pipeline automation.

Skills & Requirements

Must-have

  • Python and C++ software engineering skills
  • PyTorch ML training pipeline development
  • Model compression and quantization techniques
  • CUDA and high-performance computing optimization
  • Tool calling system architecture and implementation

Nice-to-have

  • Experience with PPO DPO GRPO optimization methods
  • Ability to read and reproduce recent research papers
  • Strong experimental discipline in algorithm implementation
  • Knowledge of KV caching and decoding-time optimizations

Key Requirements

  • Fluency in Chinese (Mandarin) and English
  • Hands-on experience with model efficiency and inference optimization
  • Proficiency in CUDA and performance tuning

Work Rights

Not specified

Tailored Resume

Cover Letter