3: Edge AI Deployment Engineer (Combined Hardware + Network)

Role Title: Edge AI Deployment Engineer — Hardware Assembly, Networking & Site Commissioning

Experience Level: 3+ Years

Location: Mysuru, Karnataka (Base Office with 30–40% Field Travel)

Employment Type: Full-time, Permanent

About the Role

We are seeking an adaptable Edge AI Deployment Engineer who combines hands-on hardware assembly expertise with field networking and site commissioning capabilities. This hybrid role covers the complete product lifecycle of our CK-AI Box platform: from barebone mini PC assembly, thermal mounting, and OS provisioning at our Mysuru facility, to on-site network configuration, multi-camera integration, and client handover across customer installations throughout India.

In this role, you will bridge the gap between physical hardware builds, enterprise IP networking, and live edge AI computer vision inference. You will be equally comfortable with a torque screwdriver assembling an industrial mini PC, working in a Linux terminal configuring dual-NIC isolation and FRP reverse tunnels, and calibrating on-site IP cameras alongside client IT and security leadership.

Key Responsibilities Across the 4 Lifecycle Phases Phase 1: Hardware Assembly & Staging (Mysuru Facility)Assemble industrial barebone mini PCs (Intel NUC, Mini-ITX form factors, fanless aluminum chassis) under strict ESD-safe conditions.Install and seat SO-DIMM DDR4/DDR5 RAM, high-speed NVMe M.2 SSDs, and specialized AI accelerators (Hailo-8 / Hailo-8L via M.2 or PCIe carrier cards).Apply thermal interface materials (pastes, gap pads) to CPU, storage, and NPU contact surfaces to ensure proper heat dissipation to the chassis.

Route internal RF pigtails to exterior chassis SMA antenna mounts, adhering to bend-radius limits.Flash standardized BIOS/UEFI revisions; configure "Restore on AC Power Loss" to Power On, configure boot order, and enable virtualization flags (VT-x/VT-d).Provision systems with Ubuntu Server LTS, install Docker container runtimes, configure Hailo RT drivers, and deploy the CK-AI Box software stack.Execute a mandatory 4-hour burn-in stress suite (stress-ng, memtest86, smartctl, sustained NPU inference loops, dual-LAN iperf3 tests), documenting QA sign-off per unit.Phase 2: Network Architecture Design & Pre-Deployment StagingReview pre-deployment site survey data: analyze camera inventories, available switch capacity, cable run lengths, and available electrical/UPS power.Calculate site bandwidth requirements for setups with 16 to 64+ cameras,



balancing recording quality with edge AI sub-stream processing constraints.Calculate PoE/PoE+ switch power requirements across IEEE 802.3af/at/bt specifications to ensure adequate power margins during night-vision IR operation.Configure Fast Reverse Proxy (frpc) client daemons on each unit, establishing unique port-forwarding mappings back to central management servers (frps) for secure remote administrative access without inbound port forwarding.Maintain pre-deployment asset registries: link serial numbers, MAC addresses, planned static IP pools, and FRP remote connection links.Phase 3: In office Deployment & System Commissioning (Client Sites)Mount and install the CK-AI Box at customer premises; establish secure, surge-protected power connections and clean network patch cabling.Configure physical Layer 2/3 managed switches: set up 802.1Q tagged VLANs to isolate surveillance camera feeds from client management networks.

Configure dual-NIC host isolation on the CK-AI Box:Bind NIC 1 exclusively to the private camera switch plane.Bind NIC 2 to the outbound corporate router/cellular gateway for cloud telemetry.Implement local host firewalls (ufw/iptables) to prevent unauthorized cross-network routing.Discover and provision on-site IP cameras (Hikvision, Dahua, CP Plus, Axis, Vivotek) using ONVIF profiles:Calibrate camera video streams: force standard H.264/H.265 profiles, disable proprietary compression modes (Smart Codec / H.264+) that corrupt decoding, and set GOP intervals to $1times$ or $2times$ frame rate.Extract and map RTSP sub-stream URLs to local AI inference pipelines.Validate live deep learning model inference (detection, tracking, ANPR) on live feeds, verifying alert dispatch via WhatsApp, Telegram, and dashboard webhooks.Lead client handover: demonstrate dashboard capabilities, instruct client IT teams on basic operation, and obtain signed Site Acceptance Test (SAT) documentation.Phase 4: Lifecycle Support, Maintenance &

- Process ImprovementServe as the primary technical point of contact for SLA-bound network, streaming, and hardware support tickets.Perform remote diagnostics over secure FRP/SSH tunnels: inspect system logs, evaluate driver states, monitor thermal profiles, and deploy container updates.Execute periodic preventive health checks: review NVMe drive health via SMART attributes, verify CPU/NPU temperatures, and audit stream integrity.Coordinate hardware RMA workflows: diagnose faulty components, replace parts, and feed field failure root causes back into in-house assembly and QA procedures.Technical Skills &
- QualificationsEducation:



Bachelor’s degree or Diploma in ECE, Computer Science, Information Technology, Electrical Engineering, or equivalent practical technical experience.Certifications: CCNA (Routing &
- Switching / Enterprise) or CompTIA Network+ certification preferred; hardware assembly or IPC certification is an added advantage.Hardware Integration: Hands-on experience assembling industrial mini PCs, applying thermal paste/pads, installing M.2/PCIe cards, configuring BIOS/UEFI settings, and using ESD protection equipment.Networking Protocols: Solid understanding of TCP/IP, UDP, DHCP, DNS, NAT, 802.1Q VLANs, QoS traffic prioritization, and subnet design (VLSM).Surveillance Protocols: Practical experience with ONVIF (Profiles S and T), RTSP stream addressing, and video codecs (H.264, H.265).Linux &
- Systems: Proficiency in Linux environments (Ubuntu Server LTS), including terminal commands (lspci, dmesg, netplan, systemd, smartctl), basic shell scripting, and Docker container execution.Edge AI Platforms: Familiarity with edge acceleration hardware, such as Hailo-8/8L NPUs or Google Coral Edge TPUs, including driver setup and inference verification.Remote Connectivity: Hands-on experience with Fast Reverse Proxy (FRP), SSH tunneling, OpenVPN, WireGuard, and Linux firewalls (ufw).Soft Skills: Professional client-facing demeanor, clear technical communication, systematic troubleshooting abilities, and willingness to travel 30–40% to client deployment sites across India.Career Progression PathwayCareer HorizonRole TitleCore Responsibilities &
- ScopeYear 1Edge AI Deployment EngineerOwn the full lifecycle: in-house assembly, staging, field deployment, camera tuning, and Tier-1/2 operational support.Year 2–3Senior Deployment EngineerLead multi-site deployment projects; mentor junior field engineers; manage complex enterprise and defense-sector network integrations.Year 3+Deployment Lead / Hardware ArchitectDefine hardware architecture roadmaps; evaluate next-generation silicon (Hailo, NVIDIA, Intel); direct nationwide deployment infrastructure.What We OfferComprehensive, end-to-end product ownership: build the hardware, deploy the network, and manage production edge AI systems in the field.Direct exposure to modern edge computing hardware (Hailo NPUs, Intel industrial platforms, high-throughput surveillance networks).Opportunities to deploy systems across industrial, enterprise, and infrastructure environments.Competitive compensation package reflecting the hybrid hardware and networking skill set.Dedicated allowance for professional networking certifications and hardware training.

Pay: ₹40,000.00 - ₹50,000.00 per month

Benefits

- Flexible schedule
- Paid sick time

Work Location: In person

📌 Hardware & Network Engineer (Mysuru)
🏢 Attentionkart
📍 Mysuru