Systems Software Engineer (Rust, ML Inference)

ai-coustics is building the reliability layer for Voice AI, the system that closes the gap between raw audio input and reliable machine understanding in production. By combining state-of-the-art speech and audio research with real-time, production-grade SDKs, we test, observe, and enable Voice AI systems to work in any environment. Our software is used by fast-growing Voice AI companies across Europe and the United States whose products require reliable performance at scale: call center agents, voice agents, telephony apps, and enterprise voice assistants. We believe voice will become the main interface for technology and ai-coustics is building the foundational infrastructure to make audio input reliable, measurable, and easy to deploy.

We are backed by leading early-stage investors including Connect Ventures, Partech, Inovia Capital, as well as angel investors from HuggingFace, DeepMind and Amazon with deep expertise in AI and developer infrastructure. These partners share our vision and are helping us build a world-class team operating with high levels of responsibility and velocity. We look for people who take ownership, think systemically, and want to solve challenging real-world problems in close collaboration with our customers. If you're motivated by developing technology that is used in practice, shaping an emerging category and setting a new standard for how voice AI works in the real world, you'll feel at home at ai-coustics.

Role Overview

ai-coustics is seeking a Systems Software Engineer to join our Systems team, working at the core of our real-time Audio AI SDK and inference infrastructure. In this role, you will help maintain, optimize, and expand the SDK that powers ai-coustics' speech enhancement and Voice AI products across a wide range of platforms, runtimes, and languages.

You will work primarily on our Rust-based inference and systems codebase, which underpins the Airten real-time inference engine, DSP modules, telemetry, model execution pipeline, and public SDKs used by developers worldwide. Your work will directly impact model performance, runtime efficiency, reliability, developer experience, and our ability to deploy neural audio models in latency-critical production environments.

This role sits at the intersection of systems programming, ML inference, real-time audio, and developer infrastructure. You do not need to be an ML researcher, but you should be excited about making neural networks run fast, safely, and predictably in real-world applications.

Ideal starting date: August/September

Tasks

ML Inference Engine & Runtime Development

• Design, implement, and optimize systems-level components of the ai-coustics SDK and inference runtime
• Improve the performance, memory usage, and stability of the Airten real-time inference engine
• Work on model execution, tensor operations, scheduling, streaming inference, and runtime abstractions
• Support deployment of neural audio models across CPU, WASM, and other constrained runtime environments
• Explore and integrate ideas from modern inference engines and ML runtimes such as Burn, ONNX Runtime, tract, TensorRT, or similar systems
• Help bridge the gap between research models and production-ready, low-latency inference

Audio, DSP & Real-Time ML Systems

• Develop and maintain DSP modules and supporting audio-processing infrastructure
• Optimize streaming workloads under strict latency, jitter, and memory constraints
• Build tooling to validate numerical correctness, real-time behavior, and model quality across platforms
• Collaborate with ML researchers to make models easier to export, test, benchmark, and deploy
• Contribute to model conversion and deployment workflows, including formats such as ONNX, internal model formats, or Rust-native representations

Language Bindings & Platform Support

• Maintain and expand our C API and public C library generated from our internal Rust codebase
• Improve and support SDK wrappers and bindings for C++, Python, and Rust via the public C API
• Maintain WASM and Node.js SDKs built directly from the internal Rust source
• Ensure consistent behavior, performance, and API guarantees across Linux, macOS, Windows, WASM, and embedded-adjacent environments

Testing, Reliability & Tooling

• Design, implement, and extend our testing pipeline, including unit tests, integration tests, numerical tests, and performance benchmarks
• Build tooling to validate real-time constraints, memory usage, model outputs, and cross-language consistency
• Improve CI workflows to ensure safe and fast iteration on a closed-source core with public-facing SDKs
• Create benchmarks and profiling workflows that help us understand runtime bottlenecks and performance regressions
• Improve observability and diagnostics for SDK integrations in customer environments

Documentation & Developer Experience

• Write and maintain technical documentation for SDK APIs, runtime internals, model deployment, and integration guides
• Collaborate with product and developer-facing teams to improve onboarding and usability
• Support internal teams and external developers by diagnosing SDK and inference issues and proposing robust fixes
• Contribute to API design with a focus on ergonomics, safety, portability, and long-term maintainability

Requirements

Technical Skills

• Strong experience in systems programming, ideally with Rust
• Solid understanding of C/C++ interoperability, ABIs, and FFI design
• Experience building or maintaining SDKs, libraries, inference runtimes, or developer-facing systems
• Familiarity with real-time systems, performance optimization, memory management, and profiling
• Experience writing tests and benchmarks for low-level or performance-critical code
• Comfortable working across multiple platforms such as Linux, macOS, Windows, and WASM
• Ability to reason about API design, unsafe boundaries, ownership, error handling, and long-term maintainability

ML Inference & Audio Systems

• Familiarity with ML inference runtimes or deploying neural networks in production
• Experience with model formats or inference engines such as ONNX, Burn, tract, TensorRT, TFLite, Core ML, or similar systems
• Understanding of how neural networks are represented, executed, optimized, and benchmarked
• Exposure to real-time audio constraints such as latency, jitter, buffering, streaming workloads, and deterministic processing
• Interest in making ML models portable, efficient, and reliable outside of Python research environments

Mindset & Collaboration

• Strong ownership mentality and attention to detail
• Comfortable working in a closed-source core with open SDK surfaces
• Ability to reason about trade-offs between performance, safety, portability, an