Senior Software Engineer, Go - LLM Team

About AssemblyAI

At Assembly

AI, we’re building at the forefront of Speech AI, creating powerful models for speech‑to‑text and speech understanding available through a straightforward API. With more than 200,000 developers building on our API and over 5,000 paying customers, Assembly

AI is helping unlock and support the next generation of powerful, meaningful products built with AI.

We’re seeking an exceptional Senior Software Engineer to join our LLM team. This role is focused on building and maintaining our LLM gateway service—a unified API platform that connects customers to multiple LLM providers. You’ll work on high‑impact projects that directly solve customer problems, improve their AI and agentic workflows, and ensure reliable access to the best models for their use cases.

This is a deeply customer‑focused role. You’ll work closely with our customer success team to understand customer challenges, help optimize their prompt strategies, build features that address their pain points, and ensure our service reliably delivers value. As a Senior Engineer, you’ll take ownership of significant features and integrations while mentoring more junior engineers. You should be passionate about writing clean, maintainable code, implementing comprehensive testing strategies, and building highly reliable systems in service of real customer problems.

• Partner closely with the customer success team to understand customer use cases, challenges, and integration needs
• Translate customer pain points surfaced by the CS team into technical solutions and product improvements
• Build features and tooling that directly address customer needs and improve their workflows
• Provide technical guidance and expertise to the customer success team to help them support customers effectively

• Own and deliver complete features and integrations within our LLM gateway service
• Build and maintain integrations with multiple LLM providers and AI services (OpenAI, Anthropic, Google Vertex, AWS Bedrock, etc.)
• Write clean, maintainable, well‑tested code following best practices
• Design and implement scalable, fault‑tolerant solutions with appropriate abstractions
• Proactively identify and address technical debt, reliability issues, and code quality concerns
• Participate in on‑call rotation to ensure service reliability and rapid incident response

• Conduct thorough code reviews focused on maintainability, testing, reliability, and architectural concerns
• Ensure proper test coverage across unit, integration, and end‑to‑end testing levels
• Improve code maintainability and extensibility through targeted refactoring
• Contribute to runbooks, incident postmortems, customer‑facing documentation, and knowledge sharing
• Identify bottlenecks and propose solutions to improve team velocity and system reliability

• Work closely with product, customer success, and engineering teams to deliver solutions
• Collaborate with external API providers to troubleshoot integration issues and improve reliability
• Translate ambiguous customer requirements into well‑scoped technical work
• Proactively communicate status, risks, and blockers
• Mentor junior engineers on coding standards, reliability practices, and customer‑centric thinking
• Participate in architectural discussions and contribute to technical decisions

• 5+ years of professional software engineering experience
• Strong proficiency in Go with demonstrated ability to write production‑quality code
• Hands‑on experience with LLM APIs (OpenAI, Anthropic, etc.) and understanding of prompt engineering concepts
• Experience building or working with API gateways, proxies, and rate limiting systems
• Deep experience with prompt engineering, chain‑of‑thought, few‑shot learning, and other LLM optimization techniques
• Experience building applications with llms or ai agents with cascading architectures
• Solid understanding of software engineering fundamentals:
• Data structures and algorithms
• System design and architectural patterns
• Testing strategies (unit, integration, end‑to‑end)
• Code review practices and…