About the Role
We're hiring a Telephony Infrastructure Engineer to own the voice platform behind Salient's AI agents: the SIP edge, Asterisk, the voice audio and dialing layers, and the call servers that connect borrowers to our agents. You will work directly with our AI agent engineers, telephony specialist, customer success, and carriers.
This is a hands-on ownership role. You will finish work in flight, make the platform reliable and safe to change, and prove it can handle the next order of magnitude of call volume. You will also design and implement customer telephony integrations and make them increasingly self-serve.
What You'll Own
- Finishing the Asterisk migration: cutting over remaining customers, trunks, and call flows with no regression, then retiring the legacy path.
- Reliability of the voice audio and dialing layers: SLOs, monitoring, incident response, and eliminating recurring failure modes.
- Zero-downtime call-server deployments: draining, graceful shutdown, safe rollout, and rollback without dropping live calls.
- Building tools to manage Kamailio and Asterisk reconfiguration without server restarts inside Kubernetes
- Backpressure, load shedding, and self-recovery when carriers, media, or AI dependencies degrade.
- Load testing and capacity planning that validates thousands of concurrent calls and CPS at the SIP edge.
- Voice observability: call-quality and platform-health metrics, alerting, runbooks, and postmortems.
- Security of the voice edge against scanning, toll fraud, and abuse.
- Customer telephony integrations: custom SIP trunks (e.g., Genesys, Bandwidth), call transfers, SIP header passing, and API prototypes with new contact-center platforms.
- Self-serve setup: productized trunk configurations, validation tooling, and docs that shorten time to go live.
What We're Looking For
- You've run production voice infrastructure at meaningful scale and have been on call for it.
- You know SIP, SDP, and RTP deeply and debug from traces and pcaps.
- You can debug EC2/K8s networking issues that result in deadair—and develop monitoring and testing plans to minimize deadair.
- You understand real-time media, WebRTC, and latency budgets.
- You design for backpressure, graceful degradation, and safe deploys of stateful services.
- You've built load tests that found real bottlenecks before customers did.
- You write clear design docs, runbooks, and postmortems.
- You've integrated SIP trunks with customer contact centers and carriers, including transfers and header passing.
Nice to Have
- Experience with Kamailio, Asterisk, OpenSIPS, FreeSWITCH, or rtpengine.
- Experience developing regional failover for SIP
- Experience with latency-aware multi-homing