Software Engineer (AI Platforms)
SingleStore6 months ago
Delhi, IndiaMid Level
Responsibilities
- Build and evolve backend services for agent orchestration, tool execution, retrieval/RAG pipelines, and model serving integrations.
- Design secure, tenant-aware APIs and control-plane workflows for AI platform components.
- Implement MCP-style tool discovery and integration patterns for agents, connectors, and internal services.
- Engineer for reliability and scale through latency and cost controls, rate limiting, fallbacks, rollouts, and incident readiness.
- Establish evaluation practices including offline test sets, regression detection, prompt/model/version tracking, and quality gates.
- Contribute to secure AI capabilities involving permissions, data boundaries, prompt-injection defenses, and auditability.
- Collaborate with product managers, designers, customers, and partner engineering teams, mentor junior engineers, and contribute to team practices.
Requirements
- Strong software engineering skills and experience with distributed systems using Go, Python, or similar languages.
- Experience building cloud-native services with Kubernetes, containers, service-to-service APIs, and CI/CD.
- At least 4 years of experience working on a SaaS product or production platform.
- Solid understanding of supervised learning, LLM concepts, embeddings, and vector search fundamentals.
- Strong debugging and problem-solving skills across services and infrastructure, including incident-style troubleshooting.
- Preferred: hands-on experience with AI agents, orchestration frameworks, RAG systems, reranking, grounding, evaluation, and model serving.
- Preferred: knowledge of AI security, vector databases or vector capabilities, observability stacks, and SLO-driven engineering.
Tech Stack
Categories
About SingleStore
SingleStore builds SingleStoreDB, a distributed SQL database that supports real-time transactions and analytics on the same engine for data-intensive applications. It sells enterprise licenses and a managed cloud service to developers and companies needing low-latency analytics on operational data. Founded in 2011 and headquartered in San Francisco, the company is privately held and has a majority investment from Vector Capital.