AI Software Engineer at Andromeda Robotics

Melbourne, Victoria, Australia -

Full Time

Start Date

Immediate

Expiry Date

25 Aug, 26

Salary

0.0

Posted On

27 May, 26

Experience

2 year(s) or above

Remote Job

Yes

Telecommute

Yes

Sponsor Visa

Skills

Python, Asyncio, Conversational AI, LLM Integration, Prompt Engineering, State Machines, Real-time Voice Pipelines, RAG Pipelines, Vector Databases, Edge Hardware Deployment, Telemetry & Instrumentation, Software Architecture, C++, Rust, ROS 2, Linux

Industry

Computer Hardware Manufacturing

Description

The Bigger Picture At Andromeda Robotics, we're not just imagining the future of human-robot relationships; we're building it. Abi is the first emotionally intelligent humanoid companion robot, designed to bring care, conversation, and joy to the people who need it most. Backed by investors such as Forerunner and Main Sequence and with customers already deploying Abi across aged care and healthcare, we're scaling fast, and we're doing it with an engineering-first culture that's obsessed with pushing the limits of what's possible. This is a rare opportunity to join a team working on complex, real-world AI and robotics problems that directly shape how Abi shows up in the world. Our Values Empathy – Kindness and compassion are at the heart of everything we do. Play – Play sharpens focus. It keeps us curious, fast and obsessed with the craft. Never Settle – A relentless ambition, bias toward action, and uncomfortable levels of curiosity. Tenacity – Tenacious under pressure, we assume chaos and stay in motion to adapt and progress. Unity – Different minds. Shared mission. No passengers. How You'll Make an Impact Abi’s conversational system is the integration point of the stack, and the most customer-facing engineering surface at Andromeda. We tie several systems together into a cohesive persona, interfacing with firmware, robotics software, onboard machine learning systems and cloud services to turn user input into intelligent, context-aware responses. We maintain Abi’s memories, personality and activities to build an engaging companion for our customers. We're hiring an AI Software Engineer to own meaningful parts of this stack: voice pipeline orchestration, conversational state management, memory and recall systems, and the evaluation and telemetry infrastructure that tells us whether Abi is getting better or worse in the field. This is a hands-on, production-focused role. You'll build systems that run on edge hardware in aged care homes. These are environments with ambient noise, unreliable networks, and real people who depend on Abi working reliably at 3am, not just in a demo. You'll own the instrumentation that measures how well those systems are performing, and use that data to drive what gets built next. You'll work across the full lifecycle: design, implement, instrument, deploy, observe, improve. Your work will directly shape how customers experience Abi and play a central role in the quality, trust, and capability of the company's core product. Key Responsibilities Develop, debug, and maintain production AI software systems across voice, conversational state, and memory/recall workflows. Own and improve critical parts of the intelligence stack, including turn-taking, conversational state orchestration, memory retrieval, prompt and context window management, and the output behaviours that define Abi's personality in customer interactions. Build and maintain telemetry, instrumentation, and quality metrics across the stack. Use data to identify regressions, measure improvements, and inform prioritisation, not just ship features and hope. Evaluate and integrate voice and language foundation models, making informed tradeoffs between accuracy, latency, cost, and resource consumption on constrained hardware. Improve software quality across reliability, performance, maintainability, and observability, with a bias toward stability in production over velocity in development. Build resilient systems that handle the real world gracefully: ambient noise, partial utterances, network drops, hardware variance, and unexpected power cycles. Communicate clearly, document decisions well, and take ownership of outcomes end-to-end. What We're Looking For 3+ years of software engineering experience, with meaningful time spent building and operating production AI/ML application systems (not research-only). Bonus if those systems were heavily asynchronous or stateful. A strong mental model of state handling. You can think naturally in terms of state machines, behaviour trees, or equivalent patterns, and can reason clearly about complex system lifecycle and transitions under real-world conditions. Hands-on experience with real-time voice or conversational AI frameworks (e.g. Pipecat, LiveKit Agents, Vocode, or equivalent streaming pipelines). Experience integrating and evaluating foundation models (LLMs, voice-to-voice) in production, like prompt engineering, graceful failure modes, model selection, and structured approaches to measuring production output quality. Solid Python experience, with real async/concurrency experience (asyncio, event loops, race conditions in multi-threaded systems). Comfort working on resource-constrained or edge hardware, understanding latency budgets, CPU/memory profiling, and deployment to non-cloud targets. Experience with ownership of a sub-system end to end, from creation to delivery. With evidence of learning and working through the challenges of taking ideas from the drawing board to the customer. Bonus Points Direct experience with LiveKit, Pipecat, or similar real-time voice agent frameworks. Experience with RAG pipelines, vector databases, or conversational long-term memory systems. Exposure to robotics or embedded Linux (ROS 2, Jetson/ARM, NixOS). Experience building evaluation harnesses, quality dashboards, or automated regression detection for AI systems. Track record of improving production reliability such as monitoring, alerting, incident response, SLOs. Lower-level language exposure (C++, Rust) for performance-critical paths. Experience with asynchronous or concurrent system design in production (not just async/await syntax, actual distributed state coordination). Publications, interesting side projects, unconventional experience, or other strong signals of initiative and capability. The salary for this position may vary depending on factors such as job-related knowledge, skills, and experience. The total compensation package may also include additional benefits or components based on the specific role. Details will be provided if an employment offer is made. If you're excited about this role but don't meet every requirement, that's okay — we encourage you to apply. At Andromeda Robotics, we celebrate diversity and are committed to creating an inclusive environment for all employees. Let's build the future together. We are an equal opportunity employer and value diversity at our company. We do not discriminate on the basis of race, religion, colour, national origin, gender, sexual orientation, age, marital status, veteran status, or disability status. Please note: At this time, we are generally not offering visa sponsorship for this role.

Responsibilities

Develop and maintain production AI systems focusing on voice pipelines, conversational state management, and memory recall for humanoid robots. Build telemetry and quality metrics to evaluate and improve the intelligence stack on edge hardware.