Software Engineer, AI Research & Prototyping

Software Engineer

AI is moving fast. New models, new orchestration patterns, and new evaluation techniques appear every month, and some of them would make our agents meaningfully better. We are hiring a Software Engineer to own that pipeline from idea to evidence.

You will stay close to what is emerging in the research community and the voice AI ecosystem, design experiments we can trust, and turn promising ideas into tested prototypes on real production data. Just as importantly, you will teach: every investigation you run ends in something the team can use, whether that is a benchmarked prototype, a technical deep-dive, or a clear recommendation with evidence behind it.

• Track emerging techniques in voice AI and agents orchestration, and identify which ones matter for us
• Build rapid prototypes and test them against real conversation data or within engineering development process
• Run head-to-head evaluations of models, providers, and techniques (reasoning approaches, speech models, orchestration patterns)
• Turn every investigation into a team-usable artifact: a benchmark, a written deep-dive, a tech talk, or a recommendation with evidence
• Work with platform engineers to hand off validated ideas for production implementation

Required

• 3+ years of software engineering or applied ML experience
• Strong coding skills; able to build and run your own experiments end-to-end without infrastructure support
• Hands-on experience with LLMs: prompting, evaluation, and an intuition for how model behavior changes across techniques and providers
• Experimental rigor: experience designing tests with controls, baselines, and honest measurement
• A track record of teaching or knowledge transfer in some form: teaching or TA experience, workshops, technical writing, internal tech talks, well-documented open source, or developer education
• Intellectual honesty: comfortable reporting that a promising idea did not work

Nice to Have

• Advanced degree (MS/PhD) in CS, ML, or a related field, or equivalent research experience
• Experience with voice or speech systems (STT, TTS, real-time pipelines)
• Publications, technical blog posts, or open-source work we can read
• Experience taking a prototype through to production with an engineering team
• Experience evaluating AI systems in healthcare or other high-stakes domains

• The team learns about relevant new techniques from you
• Ideas are adopted or killed based on evidence
• Validated prototypes hand off cleanly to platform engineers, with the reasoning documented
• Six months in, engineers across the team can explain why our stack makes the choices it makes
• Our AI decisions get faster and more confident because the evidence base keeps growing