Engineering Manager, AgentControl Evaluations

LaunchDarkly Group; Northern (KY); Hybrid; USD 163,000 - 224,000; Full time

Benefits offered by this job

RSUs

Health insurance

Dental insurance

Vision insurance

Mental health benefits

Job summary

LaunchDarkly is seeking an Engineering Manager for the AgentControl Evaluations team. You will guide a group building offline evaluations, playgrounds, LLM-as-judges, and annotated datasets, leveraging Go, Python, and TypeScript with AWS, CockroachDB, and Datadog.

You’ll balance people leadership with technical direction across multiple services powering this product. You'll partner with PM and Design, own delivery within timelines, and drive observability and incident response for production

Qualifications

• 3+ years of software engineering experience with at least 2 years leading a backend or infra team.
• Ability to partner with PM/Design to scope, estimate and sequence work with reliable estimates.
• Strong communication in distributed, multi-timezone teams.
• Proven track record coaching engineers and driving performance.
• Familiarity with observability practices and production incident response.
• Knowledge of AI tooling and providers (Anthropic, OpenAI, Bedrock, Gemini) and related tooling.

Responsibilities

• Lead and develop a team of engineers with coaching and career growth.
• Collaborate with PM and Design to plan roadmaps and tradeoffs.
• Own delivery of features with timelines and quality.
• Work with AI Agents, Observability, and Feature Management teams on shared infra.
• Communicate roadmaps, risks, and progress to engineering leadership.
• Establish norms that promote collaboration and reduce silos.
• Drive reliability and incident response for production systems.
• Support hiring to grow the team and raise engineering bar.

Skills

Go

Python

Typescript

People management

Team leadership

Communication

Distributed teams

Observability

Tools

AWS

Datadog

CockroachDB

Langchain

Arize

OpenAI

Anthropic

Job description

LaunchDarkly is seeking an Engineering Manager for the AgentControl Evaluations team. You will guide a group building offline evaluations, playgrounds, LLM-as-judges, and annotated datasets, leveraging Go, Python, and TypeScript with AWS, CockroachDB, and Datadog.

You’ll balance people leadership with technical direction across multiple services powering this product. You'll partner with PM and Design, own delivery within timelines, and drive observability and incident response for production