2 months ago
Pune, IndiaSenior / Staff+
Responsibilities
- Lead the end-to-end design, development, deployment, and release of AI-driven features.
- Create scalable, secure, and robust AI architectures and technical design documents.
- Fine-tune, deploy, evaluate, monitor, and optimize large language models.
- Develop AI agents using RAG, multi-agent orchestration, tool use, multi-step planning, and short- and long-term memory.
- Design and optimize vector databases, indexing algorithms, and external knowledge integrations.
- Develop and optimize prompts, evaluation workflows, observability, tracing, and guardrails for agentic systems.
- Collaborate with data scientists, engineers, and product teams to integrate AI capabilities into products and internal tools.
- Communicate complex technical ideas and results to technical and non-technical stakeholders.
Requirements
- 2–10 years of total experience with strong understanding of LLMs, transformer architecture, attention mechanisms, and hyperparameter tuning.
- Proven experience designing and building AI agents, including multi-agent orchestration, tool-use patterns, multi-step planning, and agent memory architectures.
- Hands-on experience with LangGraph, CrewAI, or AutoGen and RAG pipelines using documents, databases, or APIs.
- In-depth knowledge of vector databases and indexing algorithms, with practical experience using FAISS, Pinecone, Weaviate, or Milvus.
- Experience with agent observability, tracing, and guardrails, including Langfuse or equivalent tools.
- Proficiency in prompt engineering and evaluation of model quality, relevance, and efficiency.
- Proficiency in Python and experience with TensorFlow, PyTorch, and Hugging Face Transformers.
- Experience with data preprocessing, vectorization, and large-scale datasets.
- Bachelor’s degree in engineering or an equivalent degree.
- Preferred experience with graph-based architectures, knowledge graph embeddings, graph neural networks, custom model training, AWS, GCP, Azure, Docker, Kubernetes, programmatic advertising, RTB, ad auctions, MCP, or AI research contributions.
Benefits
- Hybrid work schedule with three days in the office and two days working remotely.
- Paternity and maternity leave.
- Healthcare insurance.
- Broadband reimbursement.
- Office kitchen with healthy snacks and drinks and catered lunches.
Tech Stack
Categories
About PubMatic
PubMatic builds a sell-side programmatic advertising platform that helps digital publishers and commerce media networks monetize display, mobile, video, and in‑app inventory by connecting them with DSPs and buyers. Its business model centers on fees from real-time ad auctions and related services, including RTB, yield optimization, and AI-driven decisioning. Founded in 2006, the company is headquartered in Redwood City, California, and is listed on NASDAQ under the ticker PUBM.