Staff Backend Engineer, Agents

🔥 14 minutes ago

Apply Now

Find Similar Remote Jobs

Receive Emails For Remote Jobs

📊 Check your resume score for this job

Improve your chances of getting an interview by checking your resume score before you apply.

Upload Resume

Synthesia

WebsiteLinkedInAll Job Openings

501 - 1000 employees

Founded 2017

🤖 Artificial Intelligence

☁️ SaaS

🤝 B2B

🔥 Funding within the last year

💰 $200M Series E - Synthesia on 2025-10

Artificial Intelligence • SaaS • B2B

<Synthesia> Synthesia is a SaaS AI video platform that enables businesses to create studio-quality videos without cameras, microphones, actors, or studios by using AI avatars and synthetic voiceovers. The platform supports 160+ languages, one-click translation/localization, an AI screen recorder, brand management, collaboration and analytics, and enterprise-grade security (SOC 2 Type II, GDPR). It’s marketed primarily to teams and enterprises for training, sales enablement, marketing, knowledge management and internal communications, helping companies scale video production while reducing time and cost.

📋 Description

• You will lead the infrastructure layer powering Synthesia's real-time video agents, building the platform that serves machine learning models at the critical path of agent interaction. • You will own the model serving infrastructure that enables 2-way agent communication, integrating multiple third-party model providers (currently OpenAI's Realtime API, exploring cascaded alternatives including STT + LLM + TTS stacks), and building the systems that ensure production reliability, low latency, and consistent quality across all agent outputs. • You will take ownership of the model infrastructure layer from idea through to production, working with loosely defined requirements to scope, prototype, and ship solutions that deliver real user impact across product teams building on top of your platform. • You will build across the stack, including: Backend infrastructure for model serving and orchestration, handling provider integrations, fallbacks, and load-shedding across multiple third-party APIs; Platform reliability systems ensuring 3-4 nines uptime, latency optimization, and graceful degradation; Tooling and monitoring to support product teams integrating agent capabilities into the studio and beyond; Evaluation frameworks for model quality, latency, and voice consistency across different providers and deployment scenarios. • You will frequently work on 0 to 1 problems, such as evaluating new model providers against current baselines, architecting the transition from third-party APIs to custom in-house model stacks, and ensuring the infrastructure can scale as agent products expand across Synthesia. • You will collaborate closely with product teams, R&D, and design to: Translate product requirements into infrastructure decisions and trade-offs; Evaluate feasibility and performance across different provider and architecture options; Support product teams with reliable, low-latency model access and clear integration patterns; Drive alignment on quality metrics and performance expectations across dependent teams. • You will own the evolution of this infrastructure as it scales from supporting current agent use cases to powering multiple product teams with potentially custom in-house models.

🎯 Requirements

• You have experience building and shipping backend infrastructure or platform systems in production environments. • You have a strong track record working with external APIs, third-party integrations, and orchestrating complex dependencies. • You are comfortable working in 0 to 1 environments, evaluating alternatives, prototyping, and iterating towards solutions that deliver user impact. • You can make pragmatic trade-offs when working with evolving requirements and multiple stakeholder teams. • You care deeply about reliability, latency, and observability—building infrastructure that works consistently at scale. • You are confident collaborating with product and technical teams, including pushing back when something is not feasible and proposing better alternatives. • You are willing to work across the stack, wherever the problem is (though this role is primarily backend-focused). • Bonus: You have experience close to machine learning systems, model serving, LLM inference, real-time APIs, or similar spaces. • Bonus: You've navigated provider dependencies or built infrastructure that abstracts over multiple third-party services. • Nice-to-have: Experience with speech/audio systems or understanding of STT/LLM/TTS pipelines.

🏖️ Benefits

• 25 days of leave + local holidays