Back End Engineer

The role The ML Ranking Platform team builds and runs the services that coordinate contextualized and personalized ranking on Reddit. We build ML infrastructure, including a scatter-gather ranking coordination service and feature indexes. We work closely with a team of MLEs to develop and produce new models and with a team of ML backend engineers that build inference, training services, and feature stores. Ranking is a core and growing part of the company, connecting users with the content they’re looking for is key to the future of Reddit. This is particularly salient right now, as our society is engaging in a deep discussion around the algorithms used to match users with content. Our team is acutely aware of this. Our team mantra is “Intention over Attention”. We are deeply concerned with helping users fulfill their intention to visit our company, not just capture their attention. Key Responsibilities - System design. This requires a good understanding of how the systems work, with particular focus on latency and scalability. - Development . RP works mainly with Go and some Python. Skills in testing, SOLID and design patterns are a must. - Systems architecture. RP makes extensive use of Kubernetes, Kafka, Redis and Postgres, and many internal and third party APIs, with focus on resilience, monitoring and alerting, and automation. Hard skills Proficiency in (Go main) and python (good to have), object oriented programming, design patterns. Proficiency in testing. - Ability to naturally write clear, unconvoluted, testable code. Experience with Kubernetes, Kafka, Redis (user level). Experience with AWS and/or Google Cloud (user level). Soft skills Strong team player Strong communication skills. Bonus points Experience with highly scalable systems. Experience with Terraform. Experience with ML systems or frontend (React) is a big bonus