Engineering Manager, AI & Data Infrastructure

Decagon•San Francisco, CA

1d•Onsite

About The Position

Decagon is seeking a hands-on Engineering Manager to lead the AI & Data Infrastructure team. This is a deeply technical player/coach role focused on the core systems that power Decagon's AI agents, including data and inference systems. The role involves leading the team responsible for streaming and batch pipelines, real-time databases, and GPU/model-serving platforms for LLM inference. The manager will stay involved in code and system design, participate in incident response, and contribute directly when needed. A key aspect of the role is leading by example in AI-assisted engineering, setting standards for using AI coding tools to improve efficiency and quality. The manager will also be responsible for hiring and developing a high-performing team, partnering with various internal teams to ensure fast and safe shipping of capabilities across different deployment environments (cloud, single-tenant, on-prem). Success requires strong people leadership, execution across concurrent commitments, and technical depth for architectural decisions.

Requirements

2+ years of engineering management experience leading high-performing data, ML, or infrastructure teams, with a strong IC background before that.
Deep technical depth in streaming/batch processing, analytical databases, or model-serving — you're comfortable dropping into the codebase and shipping a PR.
Hands-on experience operating large-scale data systems (Kafka, ClickHouse/Snowflake/BigQuery, Postgres at scale) and/or production model-serving infrastructure on GPUs.
Familiarity with cloud platforms (AWS, GCP, or Azure), Kubernetes, and infrastructure-as-code.
A track record of delivering multi-quarter data or ML infrastructure initiatives through ambiguity.
A strong point of view on AI-assisted engineering — you use the tools yourself and have opinions on where they work.
Care deeply about engineering craft, operational excellence, and cost discipline.

Nice To Haves

Experience operating LLM inference infrastructure in production — GPU capacity planning, multi-provider routing, and inference evals.
Experience with realtime analytics engines (ClickHouse, Pinot, Druid) and CDC pipelines at scale.
Experience delivering data and ML systems into single-tenant, on-prem, or air-gapped enterprise environments.
Experience building internal tooling or agents that use LLMs to accelerate engineering work.
Background in security and compliance frameworks (SOC 2, PCI DSS, FedRAMP, or similar).

Responsibilities

Build, lead, and develop a high-performing team of data and ML infrastructure engineers, including hiring, coaching, and performance management.
Own the technical strategy and roadmap for Decagon's AI & Data Infrastructure — streaming/batch data, realtime databases, and the GPU and model-serving stack powering LLM inference.
Stay hands-on: review designs and PRs with depth, lead architecture for hard problems, and contribute code when the team needs it.
Drive architecture for high-throughput data systems and low-latency inference, including multi-provider LLM routing and CDC pipelines at scale.
Set reliability, quality, and cost standards — data freshness SLOs, inference latency and availability, GPU and analytical cost discipline — and build an operating cadence that keeps the platform healthy as we scale.
Invest in developer and analyst experience — paved paths for producing and consuming data, and evals and observability for inference.
Raise the bar on AI-assisted engineering: define how your team uses AI coding tools to ship faster with higher quality, and build the workflows and guardrails that make this durable.
Partner with Research, Product Engineering, Platform, and customer-facing teams to deliver data and inference capabilities on aggressive timelines, including for enterprise deployments.