Gene Solutions, Vietnam's leading biotech company founded in 2017, pioneers genetic testing and AI-driven healthcare solutions. With a growing network of next-generation sequencing (NGS) labs across Asia and ambitious global expansion plans, we're creating an intelligent platform that provides personalized insights, risk predictions, and clinical tools to revolutionize pregnancy care. Join us in making a real difference for millions of families worldwide.
About the Role
As a AI Engineer, you will work closely with AI engineers, data scientists, clinicians, and product teams to ensure that LLM-powered applications are accurate, safe, explainable, and resilient to misuse, including prompt injection and adversarial inputs. You will play a critical role in building trustable AI systems suitable for healthcare environments.
Key Responsibilities
- Design and execute structured testing strategies to evaluate accuracy, consistency, and safety of LLM-powered features in medical workflows.
- Build and maintain automated LLM evaluation pipelines, including regression tests, benchmark datasets, and failure tracking.
- Identify, test, and mitigate LLM security risks such as prompt injection, jailbreaks, data leakage, and unsafe outputs.
- Develop and stress-test robust prompt templates to ensure stable behavior under adversarial, ambiguous, or edge-case inputs.
- Assess and reduce hallucination, bias, and clinical risk in LLM responses through quantitative and qualitative analysis.
- Collaborate with clinicians and domain experts to validate outputs against medical logic, terminology, and safety expectations.
- Implement guardrails such as input validation, output filtering, policy enforcement, and confidence signaling.
- Integrate LLM testing, validation, and safety checks into CI/CD pipelines and model release workflows.
- Document model behavior, known limitations, risks, and mitigation strategies for internal review and audit readiness.
- Support proofs-of-concept, demos, and production rollouts of trustworthy LLM applications in healthcare settings.
What We're Looking For
- At least 1 year of hands-on experience working with LLMs, applied ML systems, or AI-powered applications.
- Strong proficiency in Python and practical experience in testing, validating, or evaluating AI/ML models.
- Experience working with LLMs (commercial or open-source) and prompt-based systems.
- Solid understanding of LLM risks, including hallucination, prompt injection, jailbreaks, and unsafe outputs.
- Experience designing or executing model evaluation, red-teaming, or adversarial testing workflows.
- Familiarity with prompt engineering, prompt hardening, and robustness testing techniques.
- Exposure to or strong interest in the medical, healthcare, or life sciences domain.
- Ability to collaborate effectively with engineers, data scientists, and clinicians.
- Working knowledge of software engineering best practices (Git, testing, CI/CD, Docker).
- Strong documentation, analytical, and communication skills, with a focus on building trustworthy AI systems.
Why Join Us
- Competitive package including 13th-month salary, performance bonus, full social insurance on gross salary, private health insurance, annual health check-up, and company trips.
- Opportunity to work with large-scale real-world pregnancy and genomic datasets.
- Direct impact on improving maternal and prenatal healthcare outcomes across Asia and beyond.
- Collaborative environment where data science, engineering, and clinicalexpertise work closely together.
- A culture that values scientific rigor, curiosity, and ownership—your analysis will matter.
