Netskope
About this role
This is deep systems engineering: building and hardening the C++ client runtime at the core of Netskope's security product, the agent that runs on every protected endpoint. The posting describes it as lightweight, multi threaded and cross platform, macOS first and then Windows, and asks you to architect the distributed systems that keep it fast and resilient at scale, engineering for fault tolerance so failures degrade gracefully rather than breaking a user's session. You would own reliability, observability, security and performance of that runtime across real and varied hardware, and set the patterns and standards for the endpoint codebase. There is also a modern twist: building a secure agentic harness for safe, high performance execution, and the requirements ask for fluency with agentic coding systems. The specific ask is 10 or more years overall with 6 or more shipping and sustaining production C++.
Who this is for
What the posting requires:
- 10+ years of overall industry experience, with 6+ years shipping and sustaining production C++ (C++11 and later, including 14, 17 and 20): multi threaded, memory conscious, cross platform code.
- Solid distributed systems fundamentals: routing, failover, backpressure, graceful degradation, and the failure modes that come with them.
- Deep experience building and sustaining endpoint, client or systems software.
- Expertise with data structures, algorithms and performance profiling.
- Fluency with agentic coding systems and genuine curiosity about agent harnesses.
- A BSCS or equivalent is required, with MSCS or equivalent preferred.
What the work actually looks like:
- Build and harden the C++ client runtime at the core of the product: lightweight, multi threaded, cross platform, macOS first then Windows.
- Architect the distributed systems that keep it fast and resilient at scale, engineering for fault tolerance so failures degrade gracefully rather than breaking a session.
- Build a secure agentic harness for safe, high performance execution.
- Own reliability, observability, security and performance of the runtime across real, varied hardware.
- Set patterns and standards for the endpoint codebase.
Why this is a harder problem than it sounds: an endpoint agent runs on hardware you do not control, alongside software you did not choose, and it cannot fail loudly without breaking someone's working day. The posting's emphasis on graceful degradation and varied real hardware is the honest description of that constraint.
Levelling: the title spans Staff and Senior Staff, and Netskope's postings state that teams hire across levels with candidates individually assessed and levelled on skills and experience.
Location and working pattern: Bengaluru. Netskope describes open desk office layouts, so expect an in office team. Office days are not stated.
Honest fit guidance: the 6 years of production C++ is the filter, and it means shipped and sustained rather than studied. Modern C++ across macOS and Windows with real multi threading is a narrower skill than a general backend background, and people who have it are in demand. If that is you, this is one of the more technically serious roles on today's board.
- 10+ years of overall industry experience, with 6+ years shipping and sustaining production C++ (C++11 and later, including 14, 17 and 20): multi threaded, memory conscious, cross platform code.
- Solid distributed systems fundamentals: routing, failover, backpressure, graceful degradation, and the failure modes that come with them.
- Deep experience building and sustaining endpoint, client or systems software.
- Expertise with data structures, algorithms and performance profiling.
- Fluency with agentic coding systems and genuine curiosity about agent harnesses.
- A BSCS or equivalent is required, with MSCS or equivalent preferred.
What the work actually looks like:
- Build and harden the C++ client runtime at the core of the product: lightweight, multi threaded, cross platform, macOS first then Windows.
- Architect the distributed systems that keep it fast and resilient at scale, engineering for fault tolerance so failures degrade gracefully rather than breaking a session.
- Build a secure agentic harness for safe, high performance execution.
- Own reliability, observability, security and performance of the runtime across real, varied hardware.
- Set patterns and standards for the endpoint codebase.
Why this is a harder problem than it sounds: an endpoint agent runs on hardware you do not control, alongside software you did not choose, and it cannot fail loudly without breaking someone's working day. The posting's emphasis on graceful degradation and varied real hardware is the honest description of that constraint.
Levelling: the title spans Staff and Senior Staff, and Netskope's postings state that teams hire across levels with candidates individually assessed and levelled on skills and experience.
Location and working pattern: Bengaluru. Netskope describes open desk office layouts, so expect an in office team. Office days are not stated.
Honest fit guidance: the 6 years of production C++ is the filter, and it means shipped and sustained rather than studied. Modern C++ across macOS and Windows with real multi threading is a narrower skill than a general backend background, and people who have it are in demand. If that is you, this is one of the more technically serious roles on today's board.
Apply on company site
Opens www.netskope.com, the employer's own application page. Applying is always free.