We are seeking a highly skilled and hands-on Senior AI Engineer to join our AdTech team. In this role, you will be directly involved in transforming raw advertising text into actionable intelligence - extracting keywords, intent, and semantic meanings, as well as generating high-dimensional vector embeddings to power our CTR/CVR prediction models and ad-triggering engines. Additionally, you will build AI Agents and automated workflows to streamline engineering routines and enhance internal operational efficiency.
What you will do
- Ad Text Processing & Pipeline Engineering: Develop, optimize, and maintain scalable data pipelines to process unstructured ad content (titles, descriptions, keywords) and extract topics, intent, and semantic features.
- Vector Embedding & Feature Serving: Generate, evaluate, and fine-tune dense vector embeddings to serve as high-quality feature inputs for downstream Machine Learning models.
- Model Integration & Ad Targeting: Work closely with the AdTech team to integrate extracted features and embeddings into CTR (Click-Through Rate) and CVR (Conversion Rate) prediction pipelines, as well as real-time Semantic Matching engines.
- AI Agent Development & Workflow Automation: Hands-on design and implementation of AI Agents and automation workflows to reduce repetitive engineering tasks and optimize team operations.
- System Performance Optimization: Monitor, troubleshoot, and optimize the latency and accuracy of production of NLP pipelines and vector database operations.
What you will need
Must Have
- Experience: 3+ years of practical experience in NLP and Feature Engineering, ideally within AdTech, Search, E-commerce, or Recommendation Systems;
- NLP & Vector Embeddings: Hands-on experience with text vectorization tools/models (e.g., Sentence-Transformers, Hugging Face models, BGE, OpenAI embeddings) and entity/keyword extraction techniques;
- AdTech & Recommendation Understanding: Good understanding of how text features and embedding representations feed into CTR/CVR prediction models and vector retrieval systems;
- AI Agent Frameworks: Solid experience implementing AI Agents and LLM orchestration using tools such as LangChain, LlamaIndex, CrewAI, AutoGen;
- Tech Stack: Strong proficiency in Python, standard ML/NLP libraries (PyTorch, Hugging Face, SpaCy), and experience integrating with Vector Databases (e.g., Qdrant, Pinecone, Milvus, FAISS).
Nice To Have
- Experience with model optimization techniques (quantization, distillation, GPU inference acceleration);
- Experience building Model Context Protocol (MCP) servers or custom tool libraries for LLMs;
- Familiarity with compliance, audit trails, and logging frameworks for AI services in large-scale systems.
