Job Details

QAT Global is seeking for a highly skilled and passionate AI/ML professional to build a scalable AI system.

You will work closely with a cross-functional team to design, deploy, and optimize an advanced AI architecture for our client, based in California.

• Build and optimize multi-step and multi-agent workflows for production-grade Agent and RAG frameworks for this California based client partner.

• Develop scalable AI/ML solutions using Python, GCP, and LLM ecosystems.

• Implement RAG pipelines and document retrieval strategies. Preprocessing, chunking, enrichment, embedding, and reranking workflows.

•Design prompt templates, system instructions, and guardrails.

• Integrate agents with tools, API's, and internal services.

• Optimize latency, accuracy, and token usage.

• Create automated LLM evaluation and regression tests.

• Deploy applications on cloud-native environments.

• Work with Vector and Graph databases to enable high-performance retrieval.

• Collaborate with data, cloud, and engineering teams to deliver end-to-end solutions.

• Conduct performance tuning, architecture optimization, and continuous model improvements.

• 3+ years of Python development with strong AI/ML engineering experience.

• Experience monitoring, working knowledge of model versioning. Some exposure to drift detection.

• Experience with one of following: LangGraph, LlamaIndex, or DSPy.

• Hands-on expertise with LLM APIs. Gemini, Bedrock, Vertex AI, Claude.

• Solid understanding of prompt engineering and hallucination mitigation.

• Familiarity with cloud AI services.

• Understanding of the fundamentals of end-to-end RAG architectures.

• Understanding of Vector, Graph databases and retrieval systems such as Pinecone and Weaviate.

• Exposure to LangSmith, custom evaluation pipelines.

• Experience processing high-volume, multimodal documents and operationalizing data pipelines.

• Strong problem-solving, architectural thinking, and production deployment experience.