Назад
Company hidden
6 дней назад

AI Agent Architect (AI/Web3)

Формат работы
onsite
Тип работы
fulltime
Грейд
middle
Английский
b2
Страна
UAE/Malaysia/China
Вакансия из списка Hirify.GlobalВакансия из Hirify Global, списка международных tech-компаний
Для мэтча и отклика нужен Plus

Мэтч & Сопровод

Для мэтча с этой вакансией нужен Plus

Описание вакансии

Текст:
/
TL;DR
AI Agent Architect (AI/Web3): Building production-grade Agent architectures and workflows for trading intelligence, risk assessment, compliance monitoring, and intelligent customer services with an accent on Openclaw runtime design, multi-agent orchestration, and financial-grade reliability. Focus on designing planning and recovery mechanisms, integrating internal systems and external APIs, and establishing observability, evaluation, and regression-testing frameworks.

Location: Abu Dhabi, UAE; Hong Kong SAR; Kuala Lumpur, Malaysia. Kuala Lumpur Tech Office.

Company

hirify.global is a cryptocurrency exchange and digital financial platform serving users across more than 200 countries and regions, with products spanning trading, payments, wealth management, custody, institutional services, and Web3.

What you will do

  • Lead the architecture of AI Agent systems on the Openclaw platform, including execution models, lifecycle management, resource isolation, and multi-agent orchestration.
  • Build production-ready Agent workflows for trading intelligence, risk assessment, regulatory compliance, and intelligent customer service.
  • Develop Openclaw’s plugin ecosystem, capability registry, runtime performance, extensibility, and reuse.
  • Build tool-calling frameworks integrating trading engines, risk systems, data platforms, external APIs, and MCP-based protocols.
  • Establish observability, tracing, evaluation, cost monitoring, latency governance, quality baselines, and regression testing.
  • Define Agent engineering standards and enable the team through code reviews, technical sharing, and architectural reviews.

Requirements

  • 3+ years of AI systems development experience, including 1+ year of production experience with LLMs, RAG systems, or Agent frameworks. Hands-on Openclaw development and optimization experience, including its Agent Runtime, plugin ecosystem, and multi-agent orchestration.
  • Distributed systems design experience with microservices, Kafka or RabbitMQ, and asynchronous task scheduling; able to drive requirements analysis, architecture design, and technology selection independently.
  • Proficiency in at least one backend language: Golang, Java, or Node.js.
  • REST API design and containerized deployment skills with Docker and Kubernetes.
  • Strong understanding of Transformer architecture, Attention mechanisms, and Prompt Engineering including CoT, Few-shot, and Structured Output.
  • Experience with OpenAI, Anthropic Claude, and Google Gemini APIs, plus local deployment solutions such as Ollama and vLLM.
  • Software design fundamentals including design patterns, unit testing, and CI/CD.
  • Experience building LLM evaluation frameworks or Agent tracing pipelines with LangSmith, Phoenix, or Langfuse.

Nice to have

  • Experience combining MCTS, A*, or reinforcement learning with LLMs for planning and reasoning; familiarity with SFT or RLHF pipelines.
  • Agent deployments in Finance, Web3, Compliance, or other high-complexity verticals.
  • Open-source contributions or Agent, RAG, or LLM publications at NeurIPS, ICML, ACL, or ICLR.
  • Multimodal Agent development, including vision-language understanding and parsing PDFs, financial reports, or research papers.
  • Familiarity with MCP or experience designing and implementing Agent communication standards.

Culture & Benefits

  • Professional development support through a Study Growth Fund.
  • Regular team-building activities, workshops, and internal events.
  • Collaboration with an international team across global offices and regions.
  • Career advancement opportunities within a rapidly expanding global company.
  • Internal mobility opportunities to support long-term career development.

Будьте осторожны

Если работодатель просит войти в их систему, используя iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого. Это мошенники. Обязательно жмите «Пожаловаться» или пишите в поддержку. Подробнее в гайде →