Software Engineer, Agent Orchestration & AI Workflows

Sela; San Francisco (CA); Hybrid; USD 140,000 - 190,000; Full time

Benefits offered by this job

Top-of-market health coverage

401(k) with employer match

Unlimited PTO

Hybrid in-office SF schedule

Job summary

Sela in San Francisco is hiring a software engineer to build production-grade LLM applications and in-house agent orchestration.

You will optimize context handling, state, and workflows across conversations, channels, and human touchpoints, with a focus on performance, reliability, and privacy.

Join a fast-growing startup backed by top investors, with a hybrid in-office schedule four days a week in SF and a remote day on Wednesdays.

Qualifications

• 3–6 years of software engineering experience.
• Experience building production LLM applications, agent systems, workflow engines, or other complex backend systems.
• Strong Python skills and understanding of software design patterns.
• Experience with agent frameworks such as LangGraph, OpenAI Agents SDK, or similar.
• Strong understanding of asynchronous systems, state management, tool execution, and distributed workflows.

Responsibilities

• Own major components of Sela’s in-house agent orchestration framework, from architecture and implementation to reliability in production.
• Improve how agents manage context, state, tools, reasoning, and handoffs during real-time conversations.
• Extend agents to complete longer-horizon workflows across multiple conversations, channels, and human touchpoints.
• Build systems that let specialized agents coordinate and safely share context across parallel sessions.
• Add new modalities such as video while creating a consistent agent experience across every channel.
• Improve the evaluation, observability, reliability, and safety of agents handling hundreds of thousands of conversations per day.

Skills

Python

LLM applications

Agent systems

Distributed workflows

Asynchronous systems

Tools

LangGraph

OpenAI Agents SDK

Job description

Sela in San Francisco is hiring a software engineer to build production-grade LLM applications and in-house agent orchestration.

You will optimize context handling, state, and workflows across conversations, channels, and human touchpoints, with a focus on performance, reliability, and privacy.

Join a fast-growing startup backed by top investors, with a hybrid in-office schedule four days a week in SF and a remote day on Wednesdays.