Software Engineer, Agents

Description

The company is an AI video business whose aim is to make video creation much easier. The team works on hard creative and technical problems in generative media.

The focus is on new ways for people to work with AI that can do real creative work alongside them, including new interaction models for AI-native video tools. This involves multimodal agents that turn creative direction into edits, rendering engines that layer video and graphics, and generative video models built in-house that create new media when needed.

The main product is Captions, a workspace for generating, editing and designing video. Its underlying technology is also offered more widely: Tesseract adds native video editing to any AI agent, and developers can add the company's models to their own products through APIs.

The role centres on agentic systems that drive creative workflows for short-form video, helping users work faster, learn quickly and try new creative ideas. The agents act for users on their behalf while the team tests new models and methods that reach production.

Engineers on this team span product, machine learning and systems work, owning problems from start to finish and shaping how agents are built and shipped. The company has no formal titles or levels, but it is hiring engineers who have worked at senior or staff-plus scope.

Backers include Index Ventures, Kleiner Perkins, Sequoia Capital, Andreessen Horowitz, General Catalyst, Uncommon Projects, Kevin Systrom, Mike Krieger, Lenny Rachitsky, Antoine Martin, Julie Zhuo, Ben Rubin, Jaren Glover, SVAngel, 20VC, Ludlow Ventures, Chapter One and others.

Responsibilities

• Build and launch complete agent systems that improve creative work, weighing output quality, speed and reliability together
• Design agent structures covering how agents collect context, plan, choose tools and carry out creative tasks dependably at large scale
• Bring in leading models, mixing the company's own research with outside capabilities to power new agent features
• Track and raise agent quality in production through experiments, evaluation frameworks and feedback from real use

Requirements

• 5+ years of professional industry experience
• A record of designing and building ML systems or agent pipelines that run in production
• Strong hands-on work with context engineering, including RAG systems, token optimization and managing context at scale
• Experience building evaluation systems and agent infrastructure for context gathering, tool selection and multi-step planning
• Strong problem-solving skills and the ability to pick up new ideas fast in a field that changes rapidly
• Ability to work well in a very quick-moving environment, and to scope and deliver projects from start to finish
• You keep up with where generative media and agentic AI are heading
• You are grounded and collaborative, and willing to do what it takes for the team to succeed

Conditions

Most roles require working in person at the NYC headquarters in Union Square

Equal opportunity applies to all candidates regardless of background.

Benefits

• Medical, dental and vision coverage
• 401K with an employer match
• Commuter benefits
• Catered lunch on multiple days each week
• A dinner stipend every night you work late
• Grubhub subscription
• Health and wellness perks
• Several team offsites a year and a team event every month
• Generous paid time off
• Benefits apply to full-time employees only