Principal/Senior/Software Engineer, AI-Powered Advertising Agents - AgenticOS
Core
Design, develop, and deploy advanced AI agents leveraging RAG, vector databases, and LLMs to enhance generative AI capabilities for complex information retrieval and adaptive interactions.
Role type
Principal/Senior Software Engineer (AI Agents & Generative AI)
Builds
AI-driven features, agentic systems, and scalable AI architectures for PubMatic's advertising platform.
Domain
Digital Advertising / Generative AI / AI Agents
Deliverable
production ML models | product features
Required skills
LLMs and transformer architecture, AI agent design (multi-agent orchestration, tool-use, planning, memory), agentic frameworks (LangGraph, CrewAI, AutoGen), RAG pipelines, vector databases (FAISS, Pinecone, Weaviate, Milvus), prompt engineering, model evaluation (Evals), Python, ML libraries (TensorFlow, PyTorch, Hugging Face)
Preferred skills
Graph-based AI architectures (knowledge graphs, GNNs), small base model training and fine-tuning, cloud deployment (AWS, GCP, Azure), containerization (Docker, Kubernetes), programmatic advertising/RTB, MCP (Model Context Protocol), AI research publications
Technologies
LangGraph, CrewAI, AutoGen, FAISS, Pinecone, Weaviate, Milvus, Langfuse, TensorFlow, PyTorch, Hugging Face, Docker, Kubernetes, AWS, GCP, Azure
Responsibilities
Lead end-to-end design and deployment of AI-driven features; Spearhead technical design for scalable AI architectures; Implement and optimize LLMs for specific use cases; Develop AI agents powered by RAG systems; Design and optimize vector databases; Create and optimize prompts; Utilize evaluation frameworks to assess model performance; Integrate AI capabilities into customer-facing products; Monitor and optimize models for performance and cost efficiency.
Seniority
Senior/Principal, hands-on IC
