About the job
We are the TikTok Monetization - Growth team — building the next-generation AI Agent-powered intelligent growth platform that serves millions of advertisers and reaches billions of users every day. We are embedding LLM reasoning, multimodal perception, and autonomous decision-making directly into the full advertising lifecycle — from ad creation and creative generation to bidding strategy, attribution, and measurement, helping businesses worldwide efficiently connect with customers and achieve commercial success on TikTok. If you want to ship cutting-edge Agent technology at internet scale with real business impact, this is the place.
Responsibilities
- Architect & build AI Agent systems: Design and build autonomous Agent frameworks for ads growth and marketing scenarios, including but not limited to: Task Planning engines based on ReAct / Plan-and-Execute paradigms, a pluggable Tool-Use & Function Calling skill layer, a Memory management system fusing long-term and short-term memory (vector retrieval + conversation summarization), and a Multi-Agent Orchestration system supporting role-based task delegation and message routing
- Deeply integrate LLM capabilities: Design Prompt Engineering & Chain-of-Thought reasoning pipelines, connect foundation models with business systems via RAG (Retrieval-Augmented Generation) and Structured Output, enabling intelligent decision-making powered by massive advertiser profiles and ad performance data — helping advertisers achieve full-lifecycle growth from Registration → Activation → Scaling
- Build omni-channel personalized marketing Agents: Leveraging an Event-Driven Agent architecture, create AI-driven personalized recommendation and automation marketing solutions spanning WhatsApp, phone, email, ad placements, and more — delivering end-to-end Agentic Workflows (intent recognition → strategy generation → action execution → feedback loop)
- Enhance growth content creation & delivery experience: Leverage multimodal foundation model capabilities, combined with Agent self-evaluation and iterative optimization (Self-Reflection / Critique Loop), to dramatically improve creative asset quality, automated bidding strategy tuning, and overall delivery efficiency
- Collaborate cross-functionally: Work closely with Product, Operations, Sales, and Data Science teams to translate AI capabilities into measurable growth outcomes
Qualifications
Minimum
- MS or above in Computer Science, Artificial Intelligence, or a related field
- 1+ year of software engineering experience in the internet industry, with strong CS fundamentals (data structures, algorithms, OS, networking)
- Deep understanding of AI Agent architectures; hands-on experience with one or more of the following: Agent frameworks (LangChain / LangGraph / AutoGen / CrewAI), Tool-Use & Function Calling mechanisms, RAG pipeline construction (vector databases + Embedding + retrieval strategies), Prompt Engineering (CoT / Few-shot / ReAct Prompting)
- Strong system design skills with the ability to craft sound technical solutions for complex business problems
- Excellent teamwork and communication skills
Preferred
- Experience building ads / recommendation / big-data platforms, having led at least one large-scale project end-to-end
- Experience designing and deploying multi-agent systems (agent communication protocols, state management, fault tolerance, etc.)
- Hands-on MLOps / LLMOps experience — model serving, evaluation, and monitoring pipelines
- Active interest in frontier Agent research: ReAct, Reflexion, Tree-of-Thought, Plan-and-Execute, and beyond
- Open-source contributions or published technical blog posts in the AI/ML space
- Analytical mindset with product sense — able to drive technical decisions from a business perspective