Senior AI/ML Engineer: LLMs, RAG & On-Prem AI Systems

Orbina; Turkey; Híbrido; TRY 250.000 - 500.000; Jornada completa

Descripción de la vacante

Orbina is seeking a Senior AI/ML Engineer to design, build, and optimize AI systems with emphasis on Retrieval-Augmented Generation, on-prem deployments, and agentic workflows. You’ll shape our AI strategy and deliver production-ready ML pipelines across diverse domains.

The role requires 5+ years of production AI/ML experience, deep Python expertise, and hands-on with LLMs, vector databases, and MLOps in a hybrid environment.

Formación

• 5+ years of experience building and deploying AI/ML systems in production environments.
• Proficiency in Python and frameworks such as PyTorch, TensorFlow, or scikit-learn.
• Deep knowledge of LLMs, vector embeddings, prompt engineering, and RAG system design.

Responsabilidades

• Design, develop, and optimize AI/ML solutions with a focus on LLMs and RAG architectures.
• Fine-tune LLMs for domain-specific use cases and improve model accuracy and robustness.
• Build agentic workflows and multi-step reasoning pipelines for autonomous tasks.
• Architect and deploy on-premise and hybrid AI infrastructures (GPU clusters, inference servers).
• Optimize model inference for low latency and high throughput.
• Develop compound AI systems with multiple models and agents in coordination.
• Build orchestration flows using LangChain or similar frameworks.
• Implement continuous learning from user interactions and feedback.

Conocimientos

Python

LLMs

RAG

On-prem deployment

MLOps

SQL

Git

LangChain

Herramientas

Milvus

Weaviate

vLLM

GPU clusters

Descripción del empleo

Orbina is seeking a Senior AI/ML Engineer to design, build, and optimize AI systems with emphasis on Retrieval-Augmented Generation, on-prem deployments, and agentic workflows. You’ll shape our AI strategy and deliver production-ready ML pipelines across diverse domains.

The role requires 5+ years of production AI/ML experience, deep Python expertise, and hands-on with LLMs, vector databases, and MLOps in a hybrid environment.