About the Role

About the Team

Our engineering team builds next-generation, cloud-native bare-metal orchestration platforms for AI neo-cloud environments. We operate on a T-shaped engineering model: while each engineer brings deep expertise in a primary domain, every team member maintains practical fluency across neighboring systems. This shared foundation enables rigorous design reviews, effective cross-domain code reviews, and dependable on-call coverage. Our development is fundamentally based on open-source projects, and contributing to them is a major part of our work.

We develop primary control plane software in Rust and Go, leverage Kubernetes operators, and foster an OpenTelemetry-first operational culture.

Role Overview

We are seeking a Senior Software Engineer who combines strong software architecture principles with deep technical domain expertise in Datacenter Networking and DPU architectures. Fitting into our T-shaped engineering model, you will drive the development of control planes powering high-density, high-throughput infrastructure for AI training and inference workloads. Familiarity with automated bare-metal provisioning and lifecycle management is a strong asset.

Core Responsibilities

Control Plane Development: Design, build, and maintain production-grade control plane microservices, custom Kubernetes operators, and robust reconciliation engines in Rust and Go.
Architecture & API Design: Author clear Functional (FR) and Non-Functional Requirements (NFR), architectural specs, system sequence diagrams, and clean gRPC/Protobuf and REST API schemas.
Network Integration: Develop custom integration modules and integrations for modern network operating systems (SONiC, NVUE, Cumulus) and DPU hardware platforms.
Observability & Diagnostics: Instrument services end-to-end using OpenTelemetry (OTel) traces and metrics, performing systematic root-cause analysis across polyglot distributed systems.
Quality & Engineering Excellence: Participate in rigorous, review-gated pull request workflows across Rust, Go, and SQL codebases while ensuring high test coverage via mock-driven testing.

Qualifications

Technical Qualifications

Language Proficiency: Primary mastery of Rust (Tokio async runtime, Tonic, Axum, sqlx) and strong proficiency in Go.
Data & APIs: Advanced SQL / PostgreSQL fluency, gRPC/Protobuf contract design, and schema evolution.
Concurrency & Reliability: Strong background in async concurrency models, lock-free patterns, distributed state handling, and mock-driven testing discipline.
Datacenter Protocols & OS: Expertise in BGP, MP-BGP, EVPN, VXLAN, L3VNI, route targets, and route server design. Hands-on experience with SONiC, Cumulus Linux, and NVUE (featuring a first-class NVUE client).
DPU & Fabric Ecosystem: Deep knowledge of NVIDIA DOCA, Host-Based Networking (HBN), BlueField DPU architectures, and the DPF (DOCA Platform Framework) operator model (BFB, DPUSet, DPUNode CRDs).
High-Performance Interconnects: Understanding of InfiniBand fabrics, NVLink / NMX-M partitioning, and RoCEv2.

Requirements

Rust proficiency

Primary mastery of Rust and experience with its async runtime.

Go programming

Strong proficiency in Go for developing microservices.

Datacenter networking

Expertise in BGP, EVPN, and related protocols.

SQL knowledge

Advanced fluency in SQL and PostgreSQL for data management.

API design

Experience in designing gRPC/Protobuf and REST APIs.

Nice to Have

DPU architecture

Knowledge of BlueField DPU architectures and NVIDIA DOCA.

OpenTelemetry experience

Familiarity with OpenTelemetry for observability.

Benefits

Remote work

Flexible remote work options are available.

Open-source contributions

Opportunity to contribute to open-source projects.