SLM & VLM Engineer — Post-Training, Tool Calling & Agents
At a glance
Mid-level AI Research role at Jabil. Singapore · full-time.
Pay not stated AI Research salary
aimlroles summary, based on the employer's posting.
What you'll do
- Train SLMs and VLMs through continued pretraining and instruction tuning, with workflows for data, evaluation and model versions
- Develop compression and serving improvements that reduce inference latency and cost while increasing throughput
- Build tool-calling systems for tool discovery, routing, argument checks, result verification, fallbacks and tracing
- Use reward models and post-training optimization to improve tool reliability, model quality and alignment with intended tasks
- Automate dataset preparation and evaluate changes through benchmarks, regression checks, A/B comparisons and failure analysis
What you bring
- Strong Python and C++ skills, including experience building machine-learning workflows in PyTorch
- Practical experience improving model efficiency through methods such as distillation, quantization, pruning or serving changes
- Experience with CUDA or SIMD acceleration, plus profiling and performance tuning
- Professional communication skills in both Mandarin Chinese and English
Who this fits
This mid-level engineering and research role suits someone who can combine hands-on model development with disciplined experimentation. You’ll work with counterparts in China, so Mandarin Chinese and English communication are required.
From the employer
Job Summary
We are looking for a highly capable engineer/researcher to lead the R&D of Small Language Models (SLMs) and Vision-Language Models (VLMs) for edge / low-latency and cost-efficient production scenarios. You will own the continuous pretraining, supervised instruction tuning (SFT), and compression/distillation pipelines, and work closely with platform teams to deliver reliable, measurable improvements in inference efficiency, tool-use success rate, and overall model quality.
Key Responsibilities
1) SLM/VLM Training: Continuous Pretraining & Instruction Tuning (SFT)
Conduct continuous pretraining and SFT for SLMs and VLMs to improve task performance and domain adaptation.
Build reproducible training workflows in PyTorch, including data processing, training, evaluation, and model versioning.
2) Compression, Distillation & Edge/Low-Latency Inference Optimization
Design and implement efficient compression strategies for SLM/VLM, including knowledge distillation, pruning, and quantization-oriented training or post-training optimization.
Optimize model serving and inference for low-latency / edge scenarios by improving throughput and cost-per-token via techniques such as quantization, caching/KV optimizations, batching strategies, and decoding-time optimizations.
3) Tool Calling System: Catalog, Routing, Validation, Fallback & Observability
Architect and implement a production-grade tool calling (function/tool calling) framework:
Tool cataloging and metadata/schema design
Tool selection/routing and argument construction
Parameter validation, result verification, and safe fallback/retry strategies
Call-chain tracing, monitoring, and observability to improve success rate and ROI
4) RL & Reward Modeling for Alignment and Tool-Use Reliability
Apply post-training methods such as PPO / DPO / GRPO-like optimization and reward modeling to align the model toward objectives including:
semantic understanding
tool-use success rate
content generation quality and consistency
Support both offline and online iteration loops, including policy evaluation, regression checks, and safe deployment gating.
5) Data Pipeline Automation (Collection, Cleaning, Curation)
Design automated pipelines for data collection, filtering, cleaning, de-duplication, labeling/weak supervision, and dataset version management to continuously improve training quality.
Ensure datasets support both SFT and preference/RL style post-training.
6) Rigorous Evaluation, Testing & Iteration
Build robust evaluation mechanisms: offline benchmarks, task suites for tool-use, regression tests, and reliability metrics.
Drive rapid iteration through A/B comparisons, ablations, and failure analysis, improving both quality and efficiency over time.
Required Qualifications
Strong software engineering skills in Python and C++, including experience building ML training/evaluation pipelines in PyTorch.
Hands-on experience in model efficiency and inference optimization (e.g., distillation, quantization, pruning, serving optimization).
Experience with high-performance computing and acceleration: CUDA and/or SIMD, profiling and performance tuning.
Ability to read and reproduce key ideas from recent papers and implement algorithms with strong experimental discipline.
Ability to communicate effectively in both Chinese (Mandarin) and English as the successful person will have to liaise with the our counterparts in China.
BE AWARE OF FRAUD: When applying for a job at Jabil you will be contacted via correspondence through our official job portal with a jabil.com e-mail address; direct phone call from a member of the Jabil team; or direct e-mail with a jabil.com e-mail address. Jabil does not request payments for interviews or at any other point during the hiring process. Jabil will not ask for your personal identifying information such as a social security number, birth certificate, financial institution, driver’s license number or passport information over the phone or via e-mail. If you believe you are a victim of identity theft, contact your local police department. Any scam job listings should be reported to whatever website it was posted in.
Jabil, including its subsidiaries, is an equal opportunity employer and considers qualified applicants for employment without regard to race, color, religion, national origin, sex, sexual orientation, gender identity, age, disability, genetic information, veteran status, or any other characteristic protected by law.
Accessibility Accommodation
If you are a qualified individual with a disability, you have the right to request a reasonable accommodation if you are unable or limited in your ability to use or access Jabil.com/Careers site as a result of your disability. You can request a reasonable accommodation by sending an e-mail to Always_Accessible@Jabil.com with the nature of your request and contact information. Please do not direct any other general employment related questions to this e-mail. Please note that only those inquiries concerning a request for reasonable accommodation will be responded to.#whereyoubelong