Principal AI Agent Development Engineer/Architect

About the role

Bybit is seeking a senior engineer to lead the design and delivery of a next‑generation AI agent platform. The role involves building core orchestration, runtime, and lifecycle tooling while ensuring high performance, cost efficiency, and compliance for financial applications.

Key responsibilities

• Design and implement the Agent SDK, runtime engine, and core primitives to enable a steerable, verifiable conversational loop.
• Wrap internal systems as MCP servers/sub‑agents, handling tool discovery, dynamic registration, RBAC, rate limiting, and version management.
• Build an end‑to‑end Agent Development Lifecycle (ADLC) covering spec, evaluation, build, testing, A/B, canary releases, monitoring, and iteration.
• Own metrics for accuracy, resolution rate, cost, and latency, and create personalization optimization loops using data feedback.

Required profile

• Bachelor’s degree in Computer Science or related field with 6+ years of engineering experience, including 3+ years focused on AI/agent systems.
• Proven 0‑to‑1 delivery of production‑grade agent platforms; hands‑on experience with Go or Python.
• Deep expertise in LLM APIs, prompt engineering, ReAct patterns, and token budgeting.
• Experience with orchestration frameworks such as LangGraph, Dify, Eino, or OpenAI Agents SDK and multi‑agent collaboration.

Required skills

• Go
• Python
• LLM APIs (OpenAI, etc.)
• Prompt engineering & ReAct patterns
• LangGraph / Dify / Eino or similar orchestration frameworks
• RAG engineering (vector retrieval, hybrid search, reranking)

What we offer

• Study Growth Fund for professional development
• Regular internal events, workshops, and team‑building activities
• Global collaboration with a diverse, international team
• Opportunities for career advancement and internal mobility