Senior Software Engineer - Windows Endpoint Security (C#)

Overview

Build and own the Windows endpoint agent, a production-grade C#/.NET service that intercepts and inspects traffic on enterprise Windows machines, runs a local TLS proxy, and makes real-time decisions with on-device ONNX models.

What you'll do

• Own the Windows endpoint agent end to end as a production-grade Windows service in C#/.NET, including its installation, auto-updates, resilience, and performance footprint on the end user's machine.
• Work with the Windows Filtering Platform (WFP) and filter drivers to intercept, redirect, and inspect traffic at the right layer, choosing between a callout driver and a user-mode redirect.
• Build and harden a local proxy that terminates and inspects TLS traffic on the device, handling certificate trust, pinning exceptions, and protocol edge cases without degrading the user's browsing or application experience.
• Integrate and optimize ONNX models that run locally on the endpoint for real-time inference under tight CPU, memory, and latency budgets, with no cloud round trip per decision.
• Make the agent work alongside Zscaler, Netskope, and other SSE/SASE and EDR agents, covering proxy chaining, WFP filter priorities, certificate stores, and related conflicts.
• Debug real customer environments using ETW traces, WinDbg, and packet captures, harden the agent against crashes, network drops, and OS updates, and report reliability in numbers.

What you'll need

• 5+ years of experience building Windows desktop or system-level software in C#/.NET.
• Comfort with Windows services, the Win32 APIs, P/Invoke, and the boundary between user mode and kernel mode.
• Production experience building with the Windows Filtering Platform or filter drivers.
• Understanding of TCP/IP, DNS, HTTP/2, and TLS 1.2/1.3 sufficient to intercept them without breaking them.
• Experience building or maintaining a local proxy.
• Experience shipping an endpoint product that had to coexist with Zscaler, Netskope, or a similar SSE/SASE agent on enterprise devices.
• Experience integrating ONNX Runtime, or a similar inference engine, into a client application and tuning it for performance on real user hardware.
• Ability to walk through an endpoint agent shipped to production, including its architecture, the hardest bug chased, and what would be done differently.
• Ability to explain driver-level trade-offs to a product manager and deployment issues to a client's IT team in plain English.
• Experience working with US or Western teams.

Nice to have

• Experience with DirectML or quantized models.
• Enterprise deployment experience with MSI/MSIX, Intune or SCCM, and code signing.

Details

• Location(s): Pakistan; India; Argentina; Germany; Brazil; Romania.