Staff Engineer - Agentic Platform Team

What you will do:

• Vector Data Infrastructure: Design and optimize high-performance Vector Database architectures (e.g., Pinecone, Milvus, or Weaviate) to power efficient retrieval-augmented generation (RAG) and long-term agent memory.
• Scalable Multi-Use-Case RAG System: Design and build a production-grade, multiuse-case Retrieval-Augmented Generation (RAG) system capable of serving diverse retrieval scenarios including natural language query augmentation, AI agent context enrichment, and personalized analytics assistance. Define chunking strategies, embedding pipelines, and retrieval ranking logic that work across structured and unstructured data sources. Ensure retrieval latency, accuracy, and freshness SLAs are met at production scale across thousands of customers
• Enterprise Connectors Platform: Build and lead a production-grade connectors framework that ingests content from enterprise applications such as Slack, Jira, Workday, and Confluence to supplement query context in real time. Design connector abstractions that are extensible, schema-agnostic, and developer-friendly, enabling rapid onboarding of new data sources. Implement incremental sync, webhook-driven updates, and change-data-capture patterns to keep the knowledge base fresh,while ensuring connectors handle authentication, rate limiting, error recovery, and multi-tenant isolation securely and reliably.
• Cloud-Native Deployment: Lead the transition towards automated, GitOps-driven deployments using Argo CD to ensure reliable and scalable delivery of AI platform services.
• Advanced Observability: Design and build comprehensive observability solutions using the latest frameworks (such as Langfuse) to provide deep visibility into agent performance, traces, and interactions.
• System Scale & Reliability: Design, develop, test, maintain, monitor, and improve distributed systems and platform infrastructure to enable high-velocity iteration.
• Technical Leadership: Provide technical thought leadership to advance ThoughtSpot’s differentiation in democratizing data analytics for business users.

Required Skills and Qualifications:

• Experience: 10+ years of experience designing and implementing consumer-grade AI applications or distributed platforms.
• Platform Engineering & DevOps: Proven expertise in building scalable internal platforms and managing CI/CD pipelines with Argo CD in a Kubernetes environment.
• Vector Data Expertise: Deep understanding of Vector Data management, including indexing strategies, similarity search algorithms, and integrating vector stores into production AI workflows.
• Connectors: Experience designing connector/integration frameworks that ingest data from enterprise SaaS applications (Slack, Jira, Salesforce, Workday, etc.).
• AI Platforms & Agents: Proven track record of building AI platforms (LLM gateways) and experience with agentic architectures (e.g., LangChain, AutoGPT, or custom orchestrators). Familiarity with agentic AI patterns tool use, memory, planning, and orchestration frameworks (LangChain, LlamaIndex, etc.).
• Data pipelines: Experience with streaming/event-driven ingestion pipelines (Kafka, Kinesis, or similar).
• ML/NLP: Background in information retrieval or NLP/ML systems.
• Prior experience at a high-growth SaaS or cloud-native company scaling multi-tenant AI infrastructure
• Observability & Evaluation: Hands-on experience with modern AI observability stacks (e.g., Langfuse, Arize) and building benchmarking frameworks for GenAI applications.
• System Design: Experience with complex, high-volume distributed systems and an understanding of how to scale AI/ML infrastructure.
• Coding Proficiency: Strong coding skills in modern languages such as C++, Go, Java, or Python.
• Education: Master's/PhD degree in Computer Science, Machine Learning, Data Science, or a related field (Nice to have).