Embedded Systems Developer

• *Overview**

We are seeking a Senior Embedded Computer Vision Engineer with deep expertise in low-level C development on constrained edge devices, specifically Axis Communications cameras (M31xx / M1055-L series) using the ACAP SDK.

This role is focused on building high-performance, production-grade computer vision applications running directly on-camera, not in the cloud.

• *You will be working at the intersection of:**
• Embedded systems
• Computer vision
• CPU/Memory efficient programming

This is not a research role — we are looking for engineers who can ship reliable, optimized,

production code under real-world constraints. Applications without demonstrable C-based Embedded or edge computer vision experience will not be considered.

• *Key Responsibilities**
• *1.

Edge Computer Vision

Development**

• Develop and deploy ACAP applications in C for Axis cameras (M1055-L or M31XX class devices)
• Implement person detection, anon-reID, tracking, and counting
• Integrate with Axis Camera Application Platform (ACAP) APIs
• *2. Performance-Critical Systems Programming**
• Write highly optimized C code under strict CPU, memory, and thermal constraints
• Ensure low-latency (<100ms) processing pipelines
• Optimize for ARM architectures (proprietary Axis ARTPEC-8/9)
• *3.

Video Pipeline

Integration**

Work directly with

• H.264/H.265 streams
• VAPIX / Axis APIs
• GStreamer pipelines (where applicable)
• -Handle frame acquisition, decoding, and processing efficiently
• *4. AI / Inference Integration**

Integrate lightweight models using

• TensorFlow Lite (preferred)
• ONNX Runtime (if feasible on device)
• Axis Larod (experience strongly preferred)
• Optimize inference performance and memory usage
• **5.

Object

Detection & Tracking**

.Implement or integrate:

• YOLOv5/v8 (quantized / pruned)
• Custom lightweight detection models
• • Build tracking pipelines (e.g., SORT, DeepSORT, centroid tracking, LTSM backprop
• network based)
• **6.

Object

Detection & Tracking**

• • Very basic web frontend development; more HTML than use of a javascript library
• 7. Edge Eventing & Integration
• • Publish events via:
• -HTTP/REST
• -MQTT
• • Ensure reliable, secure, and low-bandwidth communication
• 8. Deployment & Lifecycle Management
• • Package and deploy ACAP applications using our build tools (aka Jenkins)
• • Debug directly on-device (SSH, logs, profiling tools)
• • Handle firmware compatibility and updates
• *Required Skills & Experience Core**
• Expert-level C programming (10+ years strongly preferred)
• Strong experience in embedded Linux systems
• Proven work AI vibe coding tools (Cursor or Windsurf preferred, bring your own)

Computer Vision

• Practical experience implementing:

o Object detection, tracking o Video analytics pipelines

• Experience optimizing models for edge deployment
• *Axis / ACAP (Highly Preferred)
• *Experience with:

o Axis ACAP SDK o Axis VAPIX APIs o Axis Larod framework

• Prior work deploying directly to Axis cameras is a major advantage
• *Performance Optimization
• ARM optimization (NEON, SIMD)
• Memory-constrained programming
• Profiling and performance tuning
• *AI / ML Integration
• TensorFlow Lite or similar edge inference frameworks
• Model quantization and pruning techniques
• *Nice-to-Have Skills
• Experience with GStreamer pipelines
• Experience with OpenCV (optimized usage only)
• Knowledge of RTSP / video streaming protocols
• Experience with MQTT-based architectures
• Familiarity with Kubernetes / cloud backends (for integration context)
• Experience with security on edge devices (TLS, certs, auth)
• *What Success Looks Like

Within the first 14 days, you will:

• Deploy a working ACAP application on Axis cameras
• Achieve real-time person detection + counting
• Maintain stable performance under production conditions
• Produce clean, maintainable, well-documented C code with 80% code coverage testing
• Demonstrate measurable improvements in:

o CPU utilization o Memory footprint o Inference latency

• *Engineering Culture & Expectations**

We operate with

• Production-first mindset (no prototypes shipped as final code)
• Strong emphasis on:

o Performance o Reliability o Security

• Clear ownership and accountability
• Working within an Agile SCRUM development paradigm and the use of Atlassian

Jira for the PLM

• *Engagement Details**
• Initial contract with potential for long-term extension
• Competitive hourly or monthly rate based on experience

Screening Process

• *Candidates will be evaluated via:**

1.

Technical

Interview (C + Systems + CV)

2.

Practical

Task

• o Optimize a small CV pipeline in C
• o OR implement a lightweight detection + tracking module

5.

Architecture

Discussion

• o Design a full edge → cloud pipeline for video analytics
• *Bonus Points (Stand-Out Candidates)**
• Have deployed production workloads on Axis cameras
• Have built custom CV pipelines in pure C (not Python wrappers)
• Have experience with real-world retail analytics or people counting
• Can demonstrate performance benchmarks and optimizations