indeed
Customer Success Services Support (MaaS Service and Model Optimization Direction)
Giantsage
الرياض, KSA
Full Time
Senior
Field
1 months ago
Large Language Models (LLM)Model as a Service (MaaS)Prompt EngineeringFine tuningTransformer ArchitectureTokenization
Free
Job Fit Check
Base Career helps you apply smarter for this job.
?%
Ready to ScanKey skills for this role
Large Language Models (LLM)Model as a Service (MaaS)Prompt Engineering
About the Role
We are seeking a technical expert with deep understanding of Model as a Service (MaaS) for a privatization project in Riyadh. You will be based at the client site, analyzing large model performance, optimizing model effect and cost, and handling Arabic language tasks.
Key Skills for This Role
Large Language Models (LLM)Model as a Service (MaaS)Prompt EngineeringFine tuningTransformer ArchitectureTokenization
Responsibilities
- Deeply analyze large model performance in customer's business scenario, understand and reproduce bad cases
- Provide optimization suggestions such as Prompt Engineering and Finetune; collaborate with internal algorithm team for model tuning
- Provide professional technical support and consulting on MaaS platform usage, model selection, API invocation, and cost optimization
- Monitor key performance indicators (TPM, TTFT, success rate) and analyze resource usage to provide cost effective strategies
- Utilize language advantages to accurately understand Arabic customer needs and convey them to internal team
Requirements
- Must have project experience related to large language models (LLM)
- Deep understanding of large model operation logic including Transformer architecture, Tokenization, inference process
- Practical experience in Prompt Engineering
- Familiarity with model performance evaluation metrics
- Understanding of one or more deep learning frameworks (TensorFlow, PyTorch)
- Proficient in Chinese (native level)
- Fluent in spoken and written English
- Arabic reading and writing skills
- Excellent communication and expression skills
- Able to accept 7x24 hour duty and standby arrangements
Full Job Posting
Job Overview
- Seeking a technical expert with deep understanding and practical experience in Model as a Service (MaaS) for a privatization project in Riyadh, Saudi Arabia.
- Based at client site, serving as key bridge connecting client's business needs with large model capabilities.
- Responsibilities include understanding client's large model usage, diagnosing model performance, collaborating with internal teams for optimization, and handling Arabic language tasks.
Core Responsibilities
- Model application analysis and diagnosis: Deeply analyze large model performance in inference, generation, and review stages; understand and reproduce bad cases.
- Model optimization and tuning support: Provide optimization suggestions such as Prompt Engineering and Finetune; collaborate with internal algorithm team for model tuning, especially for Arabic.
- Front line technical support and consultation: Provide support on MaaS platform usage, model selection, API invocation, and cost optimization.
- Performance and cost monitoring: Monitor key performance indicators (TPM, TTFT, success rate), analyze resource usage, and provide cost effective strategies.
- Cross cultural communication and collaboration: Accurately understand Arabic customer needs and convey them to internal team.
Qualifications
- Must have project experience related to large language models (LLM) such as model evaluation, application development, effect optimization, or technical support.
- Deep understanding of large model operation logic including Transformer architecture, Tokenization, inference process.
- Practical experience in Prompt Engineering and ability to improve model output quality by optimizing prompts.
- Familiarity with model performance evaluation metrics and experience in model effect evaluation or optimization.
- Understanding of one or more mainstream deep learning frameworks (TensorFlow, PyTorch).
- Excellent communication and expression skills.
Language and Communication
- Chinese: Proficient in native language for smooth communication with domestic R&D team.
- English: Fluent in spoken and written English for technical literature and international team communication.
- Arabic: Reading and writing skills to handle Arabic materials and understand local language nuances (core priority).
Work Location and Method
- Work Location: Riyadh, Saudi Arabia.
- Work Method: On site at client's location, closely collaborating with client's data scientists, AI engineers, and business team.
- Duty and Emergency Response: Able to accept 7x24 hour duty and standby arrangements.
Bonus Points
- Experience in handling multilingual NLP tasks, especially Arabic NLP projects.
- Practical operation or support experience in MaaS platforms (Volcano Ark, Alibaba Cloud Tongyi, Tencent Cloud Hengyuan).
- Overseas work or study background with strong cross cultural collaboration skills.
- In depth research on cost control and performance optimization for model services.
Apply for this job in 1 click
Skip the repetitive application forms
Install the Base Career Chrome Extension and autofill job applications across major job boards with your profile.
Trusted by over 500,000 job seekers on Base Career