Applied AI Engineer

The Applied AI Engineer will serve as the AI team’s technical anchor for a leading fintech organization in the United States. This onsite role in San Francisco, CA focuses on two streams: AI coding tooling and developer experience, and AI application paradigm exploration through prototypes and reference implementations.

Key Responsibilities

• Introduce AI coding tooling best practices and bring engineering depth from tools including Claude Code, Cursor, Devin, Replit Agent, and similar systems into ByLabs. Produce engineer onboarding playbooks and establish advanced-usage standards, including Skills, hooks, subagents, long-context engineering, and MCP integration.
• Set the bar for the Skill Marketplace by defining Skill design conventions. Deliver 5 to 10 reference-quality Skills as exemplars and raise team-wide Skill engineering quality through PR review.
• Prototype an AI-native engineering workflow by reimagining the spec design, implementation review, testing, and release operations lifecycle around AI tools. Produce a reference workflow that APAC teams can adopt.
• Build internal AI application prototypes for employee-facing scenarios in OpenClaw, A2UI, role-bound AI assistants, and related internal contexts. Validate streaming UX, transparent agent steps, long-context memory, multi-agent collaboration, and other emerging patterns through small frontier prototypes.
• Provide reference architecture code through runnable implementations that demonstrate how to build AI applications, with working code rather than documentation-only guidance.
• Support craft propagation across teams through one internal tech talk per month, one in-depth blog per quarter, and one onsite workshop in APAC per half year. Distill Silicon Valley AI application best practices into durable team assets.
• Participate in ByLab’s AI architecture review process as the U.S. AI team anchor, offering frontier perspective and taste-based feedback at the application layer.

Required Qualifications

• Education: Top-tier CS or EE BS or above; MS preferred.
• Experience: 6+ years of software engineering experience, including at least 2 years focused on LLM applications.
• Senior AI application engineering: Shipped AI applications from 0 to 1 in production (not demos). Must have at least one tour of duty as a core engineer at a top-tier Silicon Valley AI company or a leading AI product team, such as Anthropic, OpenAI, Meta, or similar.
• AI coding tooling fluency: Daily-driver experience with Claude Code, Cursor, Devin, or Replit Agent (at least one). Ability to articulate how it is implemented, including tradeoffs across Skills, hooks, subagents, MCP, and long-context engineering.
• Agent application engineering depth: Understanding of ReAct, Plan-Execute, Reflection, Multi-Agent, and Orchestrator-Worker tradeoffs. Hands-on experience with LangGraph, AutoGen, CrewAI, OpenAI Swarm, or in-house frameworks.
• Full-stack engineering: Strong backend experience (Go, Python, TypeScript). Mid-level or above frontend experience (React, Next.js). Comfortable with end-to-end streaming rendering and interaction.
• Product sense and craft: Ability to collaborate with PMs and designers and provide opinionated guidance on AI application UX patterns such as streaming, undo, context surfacing, and agent-step progress.
• Public technical influence: Demonstrated track record of public output, including blogs, talks, or OSS contributions, sufficient to represent the team’s craft externally.
• Self-direction and async collaboration: Ability to independently judge value priorities and drive prototype-to-dogfooding loops. Effective collaboration with APAC counterparts through documents, PRs, and asynchronous communication.

Technologies

• Claude Code, Cursor, Devin, Replit Agent
• Skills, hooks, subagents, MCP, long-context engineering
• OpenClaw, A2UI
• Go, Python, TypeScript
• React, Next.js
• ReAct, Plan-Execute, Reflection
• LangGraph, AutoGen, CrewAI, OpenAI Swarm
• OpenTelemetry GenAI, LangChain, LlamaIndex, Vercel AI SDK
• OpenTelemetry, Claude Code MCP

Nice to Have

• Open-source contributions to LangChain, LlamaIndex, AutoGen, CrewAI, Vercel AI SDK, Claude Code MCP, OpenTelemetry GenAI, or similar projects.
• Experience translating research outputs (papers, blog posts) into production application patterns.
• SaaS B2B product 0 to 1 experience.
• Experience productizing multilingual (English and Chinese) applications.
• AI experience in the financial or crypto domain.

Role Overview

This is a U.S.-anchored position with two core streams. The role leads AI coding tooling and developer experience efforts while also driving AI application paradigm exploration and dogfooding through internal prototypes.