Member of Technical Staff (Search Core, Indexing & Data Pipeline Stream)
At a glance
Senior Software Engineering role at Perplexity. Belgrade · full-time.
Pay not stated Software Engineering salary
techroles summary, based on the employer's posting.
What you'll do
- Build search engine foundations, including index pipelines, storage formats, retrieval methods and ranking for billions of pages
- Create streaming and batch workflows that construct search indexes in high-load environments
- Tune systems close to the hardware to improve performance, including work on concurrency and multithreading
- Optimize system behavior through low-level code changes and systems tuning
What you bring
- At least three years of hands-on systems programming in Rust, C++, C or a similar language
- You can own a project through development, deployment, production operation and scaling
- You know Python or another scripting language, plus a data processing system such as YT, Kafka or another listed tool
- You understand algorithms, data structures, multithreading trade-offs, distributed high-load services and Linux internals
- You have experience with low-level optimization, including memory use, cache efficiency, SIMD or profiling
Who this fits
This senior individual-contributor role suits a systems engineer who can take search infrastructure from implementation through production operations. The listed location is Belgrade, Serbia.
From the employer
Perplexity is looking for a highly skilled Expert Systems Engineer to join our Search Core team. For the indexing and data processing stream. This role is critical to building next-generation search products and technologies.
Responsibilities
Design and build core search engine components, especially indexing pipelines, data storage formats that operate at the scale of billions of pages
Develop streaming and batch data processing with YT systems for search index construction in a high-load environment
Push the limits of hardware performance through low-level optimizations and systems-level tuning
Tackle challenges in multithreading, concurrency, and system-level optimization
Design and build core search engine components, including indexing pipelines, retrieval algorithms, and ranking systems that operate at the scale of billions of pages
Develop streaming and batch data processing systems for search index construction in a high-load environment
Push the limits of hardware performance through low-level optimizations and systems-level tuning
Tackle challenges in multithreading, concurrency, and system-level optimization
Qualifications
3+ years of hands-on experience in systems programming (Rust, C++, C, or similar)
Ownership of full project lifecycle you don't just write a fast inner loop, you care about how the system is built, deployed, operated, and scaled in production
Knowledge of Python or other scripting languages
Knowledge of Data processing systems (YT/Kafka/Hadoop)
Passion for writing clean, efficient, and scalable systems-level code
Strong knowledge of algorithms and data structures, and the ability to apply them effectively
Deep understanding of multithreading, including various approaches, challenges, and trade-offs
Experience building high-load, distributed, and hardware-adjacent services
Solid understanding of Linux internals (syscalls, networking stack, memory model, kernel tuning)
Familiarity with low-level optimization techniques (memory management, cache efficiency, SIMD, profiling)
Nice to have
Experience developing core components of search engines, databases, or information retrieval systems
Understanding of search fundamentals: indexing, query parsing, ranking, and relevance
Experience with trading systems or other latency-sensitive real-time systems
Familiarity with cloud services, Kubernetes, and AWS infrastructure