Junior AI & Evaluation Engineer
Confidential عمان - الأردن تاريخ النشر:2025/11/18 سجلي الدخول للتقدمResponsibilities
- Collaborate with the AI Product team to deliver high-quality, scope-aligned features that meet product acceptance criteria.
- Own end-to-end delivery: define problems, write documentation, design prompts/RAG/tools, and ship features with clear success metrics.
- Build and maintain RAG pipelines, including chunking/embedding, hybrid retrieval (keyword + dense), reranking, vector indexing, and enforcing tenant isolation.
- Implement safe and reliable tool/function calling into internal APIs with proper guardrails, monitoring, and tracing.
- Develop golden test sets and regression suites; run offline/online evaluations for groundedness, factuality, safety, latency, and cost.
- Monitor retrieval performance (recall@k, MRR/NDCG, coverage, drift) and ensure launch readiness based on retrieval and answer quality.
- Uphold safety and compliance by mitigating injections/jailbreaks, applying policy filters, maintaining audit logs, and ensuring privacy alignment.
- Instrument token usage, latency, and caching strategies; implement semantic/fallback caching; manage vendor SLAs and cost optimization.
- Support LLMOps processes: version prompts/datasets/configurations, maintain CI/CD pipelines, run canary/shadow tests, and enforce “evals-as-gates.”
- Produce design documents and model cards; build reusable components, SDKs, and templates; mentor junior team members.
Qualifications
- Bachelor’s degree in computer science, Information Systems, or a related field.
- 3+ years of experience building production-grade AI/ML backend features, including recent LLM-based applications.
- Experience working at a leading MENA company or a high-bar startup.
- Native-level proficiency in Arabic and English, both written and spoken.
- Strong technical stack: Python and/or Node.js, AWS/GCP, containers, IaC, and modern CI/CD pipelines.
- Hands-on experience with LLM productization (prompt engineering, tool/function calling, RAG architectures).
- Strong evaluation capabilities: test-set creation, offline metrics, online A/B testing, regression suites, and quality dashboards.
- Familiarity with LLM and embedding models, as well as frameworks like LangChain, LlamaIndex, and Haystack.
- Experience with vector databases (Pinecone, Milvus, pgvector, OpenSearch), SQL/NoSQL, Redis, and data schema/versioning.
- Knowledge of governance principles: privacy, security, PDPL compliance, multi-tenant isolation, and auditability.
- Demonstrated AI fluency—actively uses AI/LLMs to enhance research, coding, testing, and documentation in a secure manner.
المهارات المطلوبة
- AI
تفاصيل الوظيفة
- الموقع عمان - الأردن
- القطاع تكنولوجيا المعلومات والاتصالات
- نوع الوظيفة دوام كامل
- الدرجة العلمية بكالوريوس
- الخبرات 3+
- الجنسية غير محدد
وظائف مماثلة
Confidential
الأردنConfidential Company المزيد
المزيد من الوظائف الشاغرة من صاحب العمل
مشاركة هذه الوظيفة
نحن نستخدم ملفات تعريف الارتباط لضمان حسن سير عمل موقعنا. للحصول على تجربة زيارة محسنة ، نستخدم منتجات التحليل. يتم استخدامها عندما توافق على "الإحصائيات". بيان الخصوصية
| تحليلات جوجل | إنشاء بيانات إحصائية |
| الهدف | _ga, _gat, _gid, _gali |