Software Development Engineer in Test (SDET)_FedGPT

FedGPT is Taiwan AILabs' in-house enterprise-grade on-premises LLM deployment platform, enabling government agencies, healthcare institutions, financial organizations, and academic bodies to adopt generative AI in fully air-gapped environments. Our customers include medical centers, research universities, and public sector organizations that demand the highest standards of system stability, reproducibility, and security.

【About the Role】
We are looking for an SDET to build and scale FedGPT's automated testing infrastructure, covering end-to-end quality assurance from low-level infrastructure (Kubernetes, GPU nodes, inference engines) to upper-layer applications (RAG pipeline, Web UI, API gateway). This is not a traditional manual QA role — you will solve quality problems with code, working as an engineer in close collaboration with SRE, FAE, and ML teams.

【Responsibilities】
1.Design and implement FedGPT's automated testing framework, covering unit, integration, E2E, performance, and regression testing
2.Build LLM-specific validation mechanisms: model output quality regression testing, inference latency and throughput benchmarking (vLLM ) and KV cache behavior validation
3.Collaborate with the SRE team to integrate tests into the GitLab CI/CD pipeline, and execute smoke tests and acceptance tests as part of air-gapped deployment workflows
4.Build pre-deployment acceptance test suites to help FAE quickly validate system health on customer sites
5.Track and quantify quality metrics (escaped defect rate, test coverage, MTTR), and propose improvements to the engineering team
6.Participate in code reviews with R&D, providing feedback from a testability perspective

【Essential Qualifications】
1. 2+ years of experience in software development or test automation
2.Proficient in Python, with the ability to independently design and maintain mid-to-large-scale test frameworks
3.Familiarity with Linux system operations, shell scripting, Docker, and basic Kubernetes concepts
4.Hands-on experience with CI/CD pipelines (GitLab CI, Jenkins, or GitHub Actions)
5.API testing experience (REST, gRPC), familiar with tools such as Postman, httpx, or equivalents

【Preferred Qualifications】
1.Experience with, or a strong willingness to learn, testing LLM / ML systems; familiarity 2.with model evaluation metrics or LLM-as-a-judge frameworks
3.Experience with inference engines (vLLM) or GPU monitoring tools (nvidia-smi, DCGM)
4.Experience with on-premises or air-gapped environment deployments
5.Familiarity with performance testing tools (Locust, k6, JMeter)
6.Familiarity with security testing or compliance standards (ISO 27001, HIPAA)

【What We Offer】
1.Cross-functional collaboration with ML, Dev, SRE, and FAE teams, with impact spanning platform architecture and customer delivery
2.A high-autonomy engineering culture that encourages solving repetitive problems through tooling and automation