N

Staff Software Engineer (Agentic Search) - Runtime

Nebius·Zürich·16.10.2026

 careers.nebius.com

 
 
full_time80–100%
Job written in
English
Location
Zürich
Work type
On-site
Type
Full-time

 

Nebius is a Nasdaq‑listed company that builds a full‑stack AI cloud platform, enabling developers and enterprises to move from data handling and model training to production deployment without the expense of maintaining large in‑house AI infrastructure. The firm's engineering‑focused culture tackles challenges across GPU orchestration, inference optimisation, compute, storage, networking and applied AI. Headquartered in Amsterdam with research centres throughout Europe, the UK, North America and Israel, Nebius employs over 1,500 staff, many of whom specialise in hardware, software and AI research. The opening is for a Staff Software Engineer dedicated to the runtime layer of an agent‑centric search engine. The role centres on constructing ultra‑fast, high‑throughput services that answer search queries in real time. Daily work involves shaping core runtime components, refining request pipelines, from query parsing through retrieval coordination to response composition, while respecting tight latency limits. The engineer will tune CPU, memory and data‑access patterns, embed observability hooks such as structured logs, metrics and traces, and continuously monitor throughput, latency and resource consumption to drive performance and cost gains. Close partnership with indexing and machine‑learning groups is required to keep retrieval and ranking logic separate from the core system, and to support controlled experiments and rigorous benchmarking. Candidates must bring at least eight years of experience building production‑grade backend services. Proven mastery of C++ or Rust in demanding, high‑traffic environments is essential, as is a track record of delivering low‑latency, user‑facing systems that sustain thousands of requests per second within strict timing budgets. A solid grasp of system‑level performance, including CPU, memory, networking and storage access, is required, together with hands‑on responsibility for deploying, debugging and rolling back code in live settings. The applicant should think holistically about request flows, balance correctness, speed and development velocity, and communicate clearly across engineering, machine‑learning and product teams. Additional strengths that would be advantageous include familiarity with database engine internals or cloud platforms, experience building large‑scale web APIs, work on performance‑critical domains such as trading or real‑time pipelines, and expertise in low‑level hardware optimisation. Contributions to open‑source projects, participation in competitive programming or capture‑the‑flag contests, involvement in elite technical programmes like SHAD, and a history of conference talks or technical papers are also valued. The position offers competitive pay, opportunities for professional growth, a flexible and ownership‑driven environment, and the chance to influence impactful AI initiatives within an international, collaborative team. What the role asks for: - 8+ years of production backend engineering experience - Proficiency in C++ or Rust for high‑load services - Experience delivering low‑latency user‑facing systems at thousands of RPS - Deep knowledge of CPU, memory, networking, and data‑access performance - Hands‑on operation of code in production, including deployment and incident handling - End‑to‑end perspective on request processing pipelines - Ability to trade off correctness, latency, and speed pragmatically - Strong cross‑team communication with engineering, ML, and product groups - Experience with DBMS internals or cloud infrastructure (nice‑to‑have) - Built high‑traffic web applications or large APIs (nice‑to‑have) - Background in performance‑critical domains such as trading or real‑time pipelines (nice‑to‑have) - Expertise in low‑level tuning and hardware optimisation (nice‑to‑have) - Open‑source contributions or active community participation (nice‑to‑have) - Competitive programming or capture‑the‑flag experience (nice‑to‑have) - Participation in advanced technical programmes like SHAD (nice‑to‑have) - Record of conference talks or technical publications (nice‑to‑have)

 

 

 

 

 

Kostenlos

Software Engineer: Lebenslauf-Vorlage

Bewirb dich mit einem Lebenslauf im Schweizer Aufbau — mit Beispieltext und den Anforderungen, die Software Engineer-Inserate am häufigsten nennen.

Lebenslauf-Vorlage Schweiz ansehen

Software Engineer: was der Markt gerade verlangt

Software Engineer-Stellen gehören zu den regelmässig ausgeschriebenen Berufen auf SwissJobs.app.

Am häufigsten genannt in Software Engineer-Inseraten:

  • Berufserfahrung12% der Inserate
  • Kundenkontakt9% der Inserate
  • Belastbar / flexibel9% der Inserate
Vorstellungsgespräch als Software Engineer: die häufigsten Fragen →

Weitersuchen

Ähnliche Jobs per E-Mail erhalten