Principal Engineer, AI Data Infrastructure

What is this role?

Principal Engineer, AI Data Infrastructure at DigitalOcean is a Hybrid, Engineering, Full-time role based in San Francisco, CA.

How do you apply?

Apply directly on DigitalOcean's careers site using the official application link below. HostList does not accept applications; we mirror listings from public ATS feeds updated weekly.

Full job description

Dive in and do the best work of your career at DigitalOcean. Journey alongside a strong community of top talent who are relentless in their drive to build the simplest scalable cloud. If you have a growth mindset, naturally like to think big and bold, and are energized by the fast-paced environment of a true industry disruptor, you’ll find your place here. We value winning together—while learning, having fun, and making a profound difference for the dreamers and builders in the world.

We are looking for a Principal Engineer to define the technical direction and architecture for AI Data Infrastructure at DigitalOcean. This role will lead the design, development, and operation of services that help AI-native applications ground, retrieve, reason over, and remember data at scale. These services will power DigitalOcean’s Agentic AI and Inference customers by providing production-grade knowledge bases, vector search, hybrid retrieval, context management, memory systems, and graph-based data infrastructure.

As a Principal Engineer, you will work across engineering, product, platform, and customer-facing teams to build foundational AI data services that are reliable, performant, scalable, cost-efficient, and simple for developers to use. You should be equally comfortable setting long-term architecture, making hard technical trade-offs, mentoring senior engineers, and going deep into system design when the business depends on getting the architecture right.

We are looking for someone who can span technical strategy and hands-on execution—someone who has strong distributed systems judgment, understands database and retrieval system internals, and can turn emerging AI infrastructure patterns into durable cloud services.

What You’ll Do

• Architect and guide the implementation of high-scale, reliable, secure AI data infrastructure services for agentic and inference workloads.

• Define the technical architecture for vector databases, knowledge bases, hybrid search, semantic search, context graphs, agent memory, and retrieval orchestration.

• Make foundational decisions on indexing, storage layout, sharding, replication, caching, query execution, ranking, consistency, latency, availability, and cost-performance trade-offs.

• Design systems that support multiple retrieval patterns, including dense vector search, keyword/BM25 search, metadata filtering, reranking, graph traversal, and context-aware retrieval.

• Build and operate managed services that customers can trust for production AI workloads, including observability, SLOs, capacity planning, backups, upgrades, failover, and disaster recovery.

• Partner with product managers and engineering leaders to translate customer needs and business priorities into a clear multi-year technical roadmap.

• Collaborate with Inference, Managed Databases, Storage, Kubernetes, App Platform, IAM, and Observability teams to ensure AI data services are deeply integrated into the DigitalOcean platform.

• Identify architectural bottlenecks, scaling risks, retrieval quality gaps, operational weaknesses, and cost inefficiencies before they become customer-impacting problems.

• Establish engineering standards, design review practices, operational mechanisms, and technical decision frameworks for AI data infrastructure.

• Mentor engineers across teams and raise the bar for architectural rigor, operational excellence, systems thinking, and customer impact.

• Stay current with advances in vector databases, retrieval-augmented generation, graph databases, memory systems, embedding models, reranking, agent frameworks, and AI data management.

Key Responsibilities

Architect and Build

• Design and evolve distributed AI data systems optimized for low latency, high recall, high availability, strong operational control, and efficient unit economics.

• Lead architecture for vector indexing and retrieval systems, including ANN algorithms, HNSW-style indexes, quantization, compression, partitioning, filtering, and recall-latency trade-offs.

• Architect knowledge base infrastructure, including ingestion, chunking, embedding generation, indexing, metadata management, retrieval, reranking, evaluation, and re-indexing workflows.

• Design context management and memory systems that enable agents to persist, retrieve, summarize, and reason over relevant state across sessions and tasks.

• Evaluate when to use vector search, lexical search, relational stores, object storage, graph databases, or purpose-built retrieval layers—and design clean integration patterns across them.

• Take a hands-on technical leadership role when needed to unblock delivery, validate architecture, or guide implementation of critical systems.

Reliability, Performance, and Scale

• Own architectural mechanisms for availability, failover, durability, capacity management, tenant isolation, cost controls, and operational safety.

• Lead performance tuning across ingestion, embedding, indexing, query serving, graph traversal, reranking, and retrieval pipelines.

• Define SLOs and operational dashboards for latency, throughput, recall quality, freshness, availability, error rates, cost, and customer-visible reliability.

• Drive automation for provisioning, upgrades, scaling, monitoring, alerting, incident response, and fleet operations.

• Build systems that scale from small developer workloads to large production AI applications with billions of objects, high-dimensional vectors, high query volume, and strict latency expectations.

Technical Leadership

• Set the technical vision for AI Data Infrastructure and influence architecture across multiple teams.

• Lead design reviews and author technical proposals that clarify trade-offs, risks, sequencing, and long-term platform implications.

• Establish standards for service design, APIs, data modeling, observability, operational readiness, testing, and production excellence.

• Mentor senior and staff engineers, helping them make better architectural decisions and operate with higher technical judgment.

• Create a culture where engineers understand not only how a system works, but why the design is correct for the customer and business.

Cross-functional Collaboration

• Work with product, engineering, design, sales engineering, support, and go-to-market teams to understand customer problems and convert them into scalable platform capabilities.

• Partner with customer-facing teams on architecture patterns for AI-native applications, retrieval-augmented generation, agentic workflows, and enterprise knowledge systems.

• Translate complex technical concepts into clear guidance for executives, product leaders, engineering teams, and customers.

• Help define migration and adoption paths for customers moving from self-managed vector databases, custom RAG pipelines, fragmented knowledge stores, or prototype agent memory systems to DigitalOcean-managed services.

Innovation and Future Roadmap

• Research and evaluate emerging technologies in vector databases, graph databases, AI memory, context engineering, retrieval evaluation, multimodal indexing, and agent data infrastructure.

• Identify which capabilities DigitalOcean should build, partner for, or integrate from open source.

• Build durable platform primitives rather than one-off features, ensuring DigitalOcean’s AI data services remain simple, composable, open, and cost-effective.

• Drive the evolution from basic retrieval infrastructure toward intelligent data systems that help agents learn, remember, and improve over time.

Key Metrics

• Availability, latency, durability, and operational health of AI data services.

• Retrieval quality, freshness, recall, precision, and reranking effectiveness.

• Time to ingest, index, re-index, and make customer data queryable.

• Cost efficiency across storage, memory, compute, indexing, and query serving.

• Customer adoption of knowledge bases, vector search, hybrid retrieval, and agent memory capabilities.

• Engineer