RocketHuntSmart job search
Reach out directly about this role
By city
3 years
Experience
Full-time
Employment
Hybrid, Onsite
Work Format
Senior
Grade
Backend
Specialization
IT & Tech
Industry
Corporation
Company Type
Senior Golang Developer at Temporal
Temporal at Yandex Just as in global IT, numerous tasks arise at Yandex for managing complex processes: service orchestration, long-running operations, integration of ML/AI pipelines, coordination of data building, and more. For such cases, reliable tools with scalable architecture and flexible integration are needed.
Temporal has excellent prospects of becoming a universal technology for solving many classes of problems — due to its maturity, support for different programming languages, vibrant open source, and extensive community. It hasn't happened yet, but we have come a long way: under our management, there are already seventy production clusters for products in almost every business unit of Yandex. Here are a few bright examples of using the technology:
Temporal in the World Temporal is an open-source engine for describing reliable and reproducible workflows directly in the code of familiar programming languages. Temporal implements stable execution of long-lived processes (orchestration) using event sourcing and state persistence. Thanks to this, applications become simpler, more reliable, and easier to scale.
Temporal is actively used in large companies, including Netflix, Snap, Nvidia, Airbnb, HashiCorp, Box Inc., Datadog, Comcast, Brex, and Coinbase. Among domestic companies, it is used at VK and, of course, at Yandex.
We are looking for a developer with a strong backend background, DevOps experience, a desire to understand modern distributed infrastructure, and to develop non-obvious engineering solutions. Your main task will be the development and growth of expertise in a powerful and popular open-source tool and its application at the scale of all of Yandex. Temporal is written in Go, but experience with TypeScript and even Rust — for SDK development — may come in handy when working with it.
Temporal + Yandex + ? >= ❤️ Hidden behind the question mark is a very large amount of work that can be described in one phrase: Managed Temporal. You can bring up Temporal locally with just one command, but to get a production installation — distributed across several data centers, with a replicated database under the hood, multiple instances of each abstraction layer with a load balancer on top, with authentication and authorization, with monitoring, alerts, and logs — you need to try really hard. And we solve this task not by creating all these entities and managing them manually, but by automating the process as much as possible.
Orchestration of AI Agents This is a completely new branch of development that we are paying little attention to so far, primarily due to a shortage of free hands, but in the global community, this topic is very hot. Temporal collaborates with projects such as Microsoft Azure, OpenAI, LangChain: to implement a platform for developing and deploying complex agent chains (multistep, human-in-the-loop, multi-agent flows) with state preservation and fault-tolerance. Examples include multimodal pipelines, multi-agent reasoning chains. At Yandex, we are developing an internal AI agent orchestration platform based on Temporal for our own agents.
Other Tasks First of all, we need to become a full-fledged managed solution, where we provide our users with all the necessary tools to implement their requests without the need to involve service developers. And for the future, we have even more grandiose plans: advanced cluster deployment scenarios with automatic acceptance testing, improved development tools integrated with Yandex's toolset, a new generation of our persistence layer based on YDB, turning into an internal analogue of Temporal Cloud.
More about backend at Yandex — in the channel Yandex for Backend
This list of knowledge and skills to be mastered is far from complete and includes only the most immediate needs.