Backend Software Engineer

Summary

Join Descript, a company building an AI-powered platform for audio and video content creation, as a Software Engineer. You will design, build, and optimize scalable backend infrastructure for real-time media processing and AI inference. Collaborate with various teams to meet evolving needs and build cutting-edge model training infrastructure. Evaluate and integrate third-party AI solutions, and make high-impact architectural decisions. This role requires 5+ years of backend software engineering experience and expertise in cloud infrastructure, modern data storage, and backend languages. Descript offers a competitive salary, generous healthcare, catered lunches, and flexible vacation time.

Requirements

• 5+ years of professional backend software engineering experience
• Experience with cloud infrastructure (GCP, AWS), modern data storage systems (S3, GCS, PostgreSQL, Dynamo/Bigtable), Linux, and data infrastructure (Airflow, Beam, Spark)
• Expertise in modern backend languages and technologies, such as Python, Typescript/Node.js, Go, or similar
• Deep understanding of system design principles, performance optimization, and infrastructure automation (e.g., Temporal, Kubernetes, Docker)
• Solid CS fundamentals, including data structures, algorithms, databases (Postgres, BigQuery), and familiarity with monitoring and observability tools (metrics and distributed tracing)
• Strong written and verbal communication skills, along with good judgment in technical decision-making

Responsibilities

• Design, build, and optimize scalable backend infrastructure to process, store, and serve real-time media processing pipelines and AI inference workloads efficiently
• Collaborate closely with Product, AI Research, and Engineering teams to ensure our platform meets the evolving needs of advanced audio and video workflows
• Build cutting edge model training and evaluation infrastructure
• Evaluate, integrate, and optimize third-party AI vendor solutions, collaborating closely with Product Engineering and AI Research teams to enhance our AI capabilities and accelerate feature delivery
• Make high-impact architectural decisions that balance short-term goals with long-term scalability, while mentoring teammates and championing engineering best practices

Preferred Qualifications

• Experience in AI and ML technologies (Pytorch, CUDA, distributed training)
• Expertise in digital media processing, codecs, and streaming technologies (ffmpeg, WebRTC, HLS, MPEG-DASH)
• Familiarity with GPU profiling, performance tuning, and optimization techniques
• Experience contributing to or engaging with open-source communities

Benefits

• Generous healthcare package
• Catered lunches
• Flexible vacation time