About the Role
Knightscope is seeking a Principal Software Architect to own the technical architecture and delivery of the Intelligent Control Module - the core software platform that runs on the new K1, H1, and K7 autonomous security robots. ICM is not a single system - it is a collection of deeply integrated real-time pipelines: a live video pipeline handling WebRTC/RTSP ingest, GStreamer-based multi-branch processing, transcoding, recording, and S3 clip delivery; a two-way audio pipeline covering SIP/VoIP, talk-down, text-to-speech, and sub-200ms round-trip audio; and a multi-model AI detections pipeline running people, vehicle, license plate, face, and thermal detection on NVIDIA Jetson edge hardware using DeepStream and YOLO-family models. These pipelines must operate reliably and concurrently on constrained edge hardware, integrate cleanly with cloud services, and scale consistently across all three robot platforms. The Principal Software Architect owns the architecture that makes all of this work - and is the single technical authority responsible for ensuring these systems are coherent, maintainable, and production-ready.
Location Requirement: Full-time, on-site at Sunnyvale HQ (No relocation provided)
Key Responsibilities
- Own the end-to-end ICM software architecture across all repositories - defining system design, data models, API contracts, and integration patterns spanning the audio, video, and AI detections stacks across the new K1, H1, and K7 robot platforms.
- Establish and enforce architectural standards, coding patterns, and reusable reference designs that reduce duplication and increase delivery velocity across the ICM team.
- Define and maintain the interface contract between the Access Control Management system and ICM; own the Signals integration strategy and cross-platform API governance.
- Own the architectural strategy for the real-time audio and video pipeline - including WebRTC ingest, multi-stream rendering, two-way audio, and edge processing on NVIDIA Jetson hardware.
- Define the AI detections architecture: edge inference pipeline for people, vehicle, license plate, and thermal detection using DeepStream/YOLO-family models on Jetson Orin NX/Xavier.
- Drive platform commonality across the new K1, H1, and K7 - identifying shared services, reducing platform fragmentation, and establishing a unified ICM architecture roadmap.
- Own field pilot readiness gates: coordinate with hardware, DevOps, and program teams to ensure ICM software meets reliability, security, and performance standards before deployment.
- Define secure release strategy including FIPS 140-2 compliance posture, credential management, and secure OTA update delivery across the robot fleet.
- Mentor senior ICM engineers across audio/video, ML detections, and DevOps disciplines; lead design reviews and foster a culture of engineering excellence.
- Represent ICM architecture to executive leadership, enterprise customers, and external partners - translating technical decisions into business impact language.
Required Qualifications
- 10-15 years of software engineering experience, with at least 4 years in a Principal Engineer, Staff Engineer, or Architect role owning platform-level architecture across multiple product lines or teams.
- Proven track record delivering software platforms for IoT, robotics, or autonomous systems in production - ideally spanning onboard edge software and cloud-connected services across multiple hardware generations.
- Hands-on experience with real-time audio/video systems: WebRTC, GStreamer, RTSP/RTMP, SIP/VoIP, or low-latency media pipeline architecture.
- Familiarity with AI/ML inference on edge hardware: NVIDIA Jetson platforms, DeepStream, TensorRT, or equivalent edge inference frameworks.
- Deep expertise in distributed systems and API design: microservices, event-driven architecture, REST/gRPC API governance, and multi-product platform integration patterns.
- Strong proficiency in cloud-native architecture on AWS (or equivalent): ECS/EKS, Lambda, IoT Core, Greengrass, S3, IAM - and infrastructure-as-code (Terraform, CDK, or equivalent).
- Familiarity with security compliance frameworks relevant to physical security and public safety deployments (FIPS 140-2, NIST 800-53, or equivalent).
- Demonstrated ability to drive architectural alignment across multiple engineering disciplines - audio/video, ML, DevOps, and platform - without direct authority.
- Excellent written and verbal communication skills; ability to produce clear architecture decision records and present technical trade-offs to both engineers and executives.
- BS/MS in Computer Science, Software Engineering, Electrical Engineering, or related field - or equivalent professional experience.
Compensation & Benefits
- Base Salary: $190,000 - $240,000 (DOE)
- Equity: Stock options
- Benefits: Medical, dental, vision, 401(k), paid time off
- Location Requirement: Full-time, on-site at Sunnyvale HQ
