Hedra is a frontier artificial intelligence company for generative media. We're building world-leading models and beautifully simple software for turning your ideas into impact, with backing from leading VCs at a16z, Index Ventures, and Abstract Ventures. We've grown to millions of creators on our platform.
SummaryAs an Agentic Engineer, you will design, build, and scale the systems that power AI agents and agentic workflows for Hedra's transformative products. You'll architect production-grade agent infrastructure moving well beyond simple API calls. You'll work with frameworks like PydanticAI, AG-UI, and A2A, implement tool calling and multi-agent orchestration, and establish rigorous evaluation practices.
Must-Have ExperienceBuilt conversational agents or agentic systems leveraging foundation models (OpenAI, Gemini, Claude, Grok, etc.) at critical scale
Strong command of TypeScript, React, Node.js, and Python
Experience with LangChain, LangGraph, pydantic-ai, or equivalent orchestration frameworks
Solid grasp of agent architecture: tool/function calling, ReAct, multi-agent orchestration, agent loops
Hands-on experience with agent evals: trajectory evaluation, tool call accuracy, benchmarking
Experience integrating LLM-based systems into user-facing web products
Startup or 0→1 product environment experience
Production experience with conversation memory, context window management, RAG, and memory summarization
Experience with agent-frontend protocols: AG-UI, SSE streaming, tool streaming
Background in observability and monitoring for agentic systems
Experience with Kubernetes
Competitive compensation and equity
401k
Healthcare (Silver PPO Medical, Vision, Dental)
Lunch and snacks at the office
We encourage you to apply even if you don't fully meet all the listed requirements; we value potential and diverse perspectives, and your unique skills could be a great asset to our team.
Top Skills
What We Do
Hedra is an AI native platform for multimodal creation. The platform is built around their own cutting-edge proprietary video model, Character-3, which is the first multimodal model in production. Alongside Character-3, the platform also brings other leading foundation models into one ecosystem spanning generative images, video, and audio. Prosumer and enterprise users leverage Hedra to generate content ranging from viral social media to branded content marketing.
Why Work With Us
We're an early-stage team that moves very fast and is building at the leading edge of AI/Media. Every employee takes on a lot of ownership and has an opportunity to learn and grow rapidly.
Gallery
Hedra Offices
OnSite Workspace
Hedra's main office is in San Francisco and secondary hub is in New York.
%20(1).png)
