Staff Software Engineer

Lupitor; San Francisco (CA); On-site; USD 150,000 - 190,000; Full time

Job summary

Lupitor is building the infrastructure layer for enterprise voice AI, delivering on-prem and air-gapped deployments for regulated environments. You will design, implement, and iterate the systems powering live customer conversations at scale, taking ideas from concept to production and owning refinement end to end.

You will optimize the inference stack, focusing on latency, throughput, and cost, and ensure portability and operability across diverse customer environments while maintaining high

Qualifications

• Experience building and shipping production backend or infrastructure systems.
• Strong grasp of distributed systems and real-time workloads.
• Ability to work independently and own large pieces of work.
• Experience with AI-assisted development workflows to ship high-quality code fast.
• Sound technical judgment on architecture and tradeoffs.
• Clear communication with technical and non-technical teammates.

Responsibilities

• Build and ship the voice infrastructure that runs in production.
• Own the inference stack for TTS and speech-to-speech, focusing on latency, throughput, and cost.
• Deploy into on-prem and air-gapped customer infrastructure with security constraints.
• Move fast without lowering the bar on reliability, performance, and code health.

Skills

Production backend

Distributed systems

Real-time workloads

Performance engineering

Ownership

AI-assisted development

Architecture judgment

Communication

Job description

Lupitor is the infrastructure layer for enterprise voice AI. We build agents that handle real customer conversations in production, running on the customer's own infrastructure instead of the cloud. The bet underneath: open source and data sovereignty out-compete cloud AI long term, and voice is where that bet lands first. To make that work we build the inference layer ourselves - serving and caching engines for TTS and speech-to-speech - because owning the stack is what makes on-prem deployment actually fast enough to use.

What You Will Work On

Build and ship the voice infrastructure that runs in production

You will design, implement, and iterate on the systems that power live customer conversations at scale - from call orchestration and telephony to the inference path that makes real-time voice possible. You will take ideas from concept to production, owning refinement end to end.

Own the inference stack

You will work on our serving and caching engines for TTS and speech-to-speech - optimizing latency, throughput, and cost in the places where the product actually wins or loses. This means profiling, benchmarking, and making real-time systems faster under real load.

Deploy into environments that don't behave like yours

You will ship our platform into on-prem and air-gapped customer infrastructure - banks and governments with their own security, network, and operational constraints. You will think about portability, operability, and failure modes, not just happy paths.

Move quickly without lowering the bar

You will balance speed and quality, shipping fast while maintaining a strong standard for reliability, performance, and code health. Voice systems fail loudly - yours shouldn't.

What Success Looks Like

• You are viewed as a high-leverage engineer the team trusts with the hardest problems
• You ship infrastructure that runs inside regulated, production calling environments and stays up
• You take ambiguous problems - a latency spike nobody can explain, a deployment that works nowhere but on your laptop - and turn them into systems the team builds on
• Your work compounds: tooling, playbooks, and improvements that make the whole team faster

About You

• Care deeply about a high standard of craft and pushing others to do the same
• Strong attention to detail and a degree of thoroughness that stands out amongst your peers
• Pragmatic about engineering decisions, with a strong intuition for simplicity
• Curious and resourceful, facing ambiguity with a bias towards action
• Obsessed with automation and efficiency, aggressively eliminating repetitive work
• Driven by seeing your work matter in production, not in demos

Skills & Experience

• Strong experience building and shipping production backend or infrastructure systems
• Comfort with distributed systems, real-time workloads, and performance engineering
• Demonstrated ability to work independently and own large pieces of work
• Experience with modern AI-assisted development workflows to ship high-quality code fast
• Sound and pragmatic technical judgment around architecture and tradeoffs
• Clear communication with both technical and non-technical teammates

Stack

Experience using our stack is not necessary for this role, but you are expected to have worked on enterprise products at scale with a modern frontend and backend. That said, we use Go for our servers, TypeScript for our web apps (Next.js) and backend, and AWS for hosting. We use Convex on top of PostgreSQL for DB and manage our infrastructure with Terraform.