SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Dyna Robotics is building general-purpose robots powered by a proprietary embodied AI foundation model. The company is already deployed with customers across multiple industries and is backed by leading investors including CRV and First Round. The team includes engineers from Google DeepMind, Meta, and Cruise.
This role owns the technical side of network performance for Dyna's robots, which operate in two critical settings: ensuring demos never drop mid-pitch, and keeping deployed robots reliably online and controllable on customer networks that Dyna doesn't own or control. This is a systems-building role focused on designing and implementing monitoring, automated failover, and self-healing diagnostics that make connectivity reliable by default.
Key responsibilities include:
- Owning network performance for remote teleoperation: measuring and minimizing end-to-end latency and jitter, tuning video and control stream transport and prioritization, and ensuring graceful degradation (lower bitrate, prioritized control traffic, safe fallback) rather than failure when links degrade.
- Building the fleet connectivity stack: monitoring, diagnostics, and automated failover software that keeps every robot's connection healthy in production, converting dropped connections into automatic root cause analysis and recovery rather than on-site troubleshooting.
- Owning the network side of site bring-up: building tooling to assess customer networks before robot deployment, then configuring appropriate failover, QoS, and VPN setups for each site.
- Designing connectivity for real-world conditions: warehouse WiFi dead zones, cellular failover, and VPN-gated corporate networks with unchangeable firewall rules.
- Diagnosing field issues on networks outside your control, then building fixes to prevent recurrence.
- Ensuring office and demo networks are rock-solid by default.
- Partnering with engineers and researchers on network requirements for training-data upload and live model debugging.
- Being the technical lead on connectivity during live demos and field incidents.
Requirements:
- Hands-on experience debugging real-world networks: WiFi/RF troubleshooting and packet-level diagnostics (Wireshark, tcpdump, iperf)
- Deep understanding of TCP, UDP, and their behavior on lossy, high-jitter links; strong Linux networking skills (routing, traffic shaping with tc, firewalling, VPNs such as WireGuard)
- Track record of shipping production-quality software in C++, Rust, or Python (not just scripts and prototypes)
- Comfort working in the field and communicating clearly under pressure; ability to translate failures into fixes during demos and deployments
- Degree in CS, EE, or related field, or equivalent practical experience
Strongly preferred:
- Experience with real-time media or low-latency transport: WebRTC, RTP, adaptive bitrate, congestion control, jitter buffers, or UDP-based protocols like QUIC or SRT
- Experience managing connectivity for a fleet of edge devices (robots, vehicles, IoT) deployed on networks you don't own
- Hands-on work with cellular/LTE/5G failover hardware, link bonding, or multipath networking
Bonus:
- Experience with pub/sub or messaging middleware (ZeroMQ, ROS 2, DDS, or similar), especially over WiFi or WAN links
- Experience with network hardware: managed switches, access points, mesh networking, cellular routers
- Experience supporting live events, demos, or on-site customer deployments where network reliability was your responsibility
