简历智能匹配【已结束】【全职远程】AI Engineer | 3K ~ 5K ｜北美远程工作

Company Introduction

We’re an early-stage AI startup dedicated to building intelligent systems from the ground up, focusing on automating complex workflows in trust & safety, fraud detection, and compliance. Our mission is to boost human productivity by developing LLM-powered agentic applications—solutions that can analyze, reason, and act on information in real time. By transforming manual review into intelligent automation, we aim to drive efficiency, scalability, and accuracy for businesses navigating critical risk-management tasks.

Job Responsibilities

• Design and deploy end-to-end retrieval-augmented generation (RAG) pipelines, ensuring alignment with business needs in trust & safety, fraud detection, or compliance.

• Manage vector databases (e.g., Pinecone, Weaviate, FAISS) to enable high-precision semantic search and reliable contextual grounding for LLM applications.

• Build agentic memory systems that support persistent, stateful reasoning across user sessions—laying the foundation for consistent, context-aware AI interactions.

• Collaborate closely with cross-functional engineering teams to integrate intelligent agents into production systems, conducting rigorous testing to ensure stability and performance.

• Optimize LLM inference performance (e.g., via quantization, batch processing) to reduce latency and cloud infrastructure costs, without compromising output quality.

• Experiment with orchestration frameworks (LangChain, LlamaIndex, OpenAI Function Calling) to design flexible, scalable workflows for agentic applications.

• Contribute to defining the technical roadmap of next-generation AI agents, bringing innovative ideas to solve open-ended problems in risk and compliance automation.

Requirements

Mandatory Qualifications

• 3+ years of hands-on experience developing AI/ML or NLP-based systems using Python (e.g., building production-grade NLP pipelines, LLM integrations).

• Deep understanding of LLMs (e.g., GPT-4, Claude 3, LLaMA 3) and practical expertise in prompt engineering (e.g., optimizing prompts for accuracy, efficiency).

• Proven track record in building agentic applications or orchestration workflows—experience with LangChain, LlamaIndex, or similar tools is required.

• Hands-on experience with RAG pipelines and vector databases (e.g., designing data ingestion flows, optimizing search relevance).

• Knowledge of agentic memory management, context window handling, and multi-turn reasoning (e.g., designing stateful agent logic).

• Familiarity with API integrations (e.g., LLM APIs, internal service APIs), model deployment (e.g., Docker, Kubernetes), and cloud environments (AWS/GCP).

• Experience with NLP libraries (spaCy, Hugging Face Transformers, OpenAI APIs) and ability to conduct model fine-tuning experiments (e.g., LoRA, supervised fine-tuning).

• Strong problem-solving, rapid prototyping, and system design skills—ability to translate business needs into technical solutions.

Preferred Qualifications (Bonus)

1. Prior experience applying LLMs to trust & safety (e.g., content moderation), fraud detection (e.g., anomaly identification), or compliance (e.g., regulatory document analysis) use cases.

2. Excellent English listening and speaking skills (Professional Working Proficiency or above)—ability to collaborate with global teams or engage with English-speaking stakeholders effectively.