Cameron O.

My name is Cameron O'Brien and I am a hands-on fractional CTO for companies betting on voice AI. Most companies making that bet have nobody who owns the technical side. The founder is selling. The operations lead already has a full-time job. The platform demo looked great. Then it hits production and nobody is accountable. That is the job I take. I make the architecture calls, pick the platforms and vendors, and do the building myself. My name goes on the outcome. Hear my work before you message me: (423) 219-1131. The agent that answers runs on one of the stacks I deploy for clients: a custom pipeline on Google Cloud with self-hosted speech models and Twilio telephony. ---------- A recent example: Boost Health Insurance runs 10,000 to 20,000 calls a day. An enthusiastic, semi-technical COO had brought in voice AI and gotten stuck where most companies do: a great-sounding web demo that fell apart on real phone calls. Latency was over a second, callers were hanging up, and the focus had drifted to how the voice sounded instead of how the calls performed. I started by verifying their numbers, and found that 67% of the transfers their platform reported as successful were actually voicemails. No one had caught it. From there I fixed what the data showed. I rebuilt voicemail detection, then moved them to a platform with answering-machine detection at the telephony layer, so it can reliably tell a human from a machine. I built them a self-improving agentic dialer: an AI agent writes challenger scripts from call outcomes, tests them head to head with Bayesian A/B testing, and shifts call volume to the winner on its own. Their sales team runs on a legacy ViciDial system that couldn't accept clean transfers from the AI, so I built an Asterisk SIP bridge that connects modern voice AI to legacy telephony, passing each caller's number through intact so their dialer routes it correctly. I also negotiated a HIPAA agreement with the platform vendor that is normally reserved for $160K enterprise contracts. From their review: "We ended up doubling our transfer rate ahead of target... He communicates clearly, gives helpful context around any issue that comes up, responds quickly, and stays solutions-focused." ---------- What I take ownership of: • Platform selection. Bland, Retell, Vapi, or fully custom (Pipecat, Deepgram, Twilio, GCP). • Making it work in production. Latency, transfer validation, voicemail detection, A/B testing, live reporting. • The systems around the agent. Multi-tenant logins, dashboards, admin panels, CRM integrations, call analytics. • HIPAA. I negotiate and lock in the BAAs your deployment needs. • The build itself. I write the code and stand behind what ships. A recent advisory example: the team building Foundry, a job management platform for home-service contractors, brought me in to scope where AI added real value and where simpler rules-based structure was the better call. From their review: "His guidance reduced build risk, clarified MVP priorities, and helped our team move faster with more confidence. He's sharp, pragmatic, and focused on shipping cleanly rather than over-engineering." I work on monthly retainers, with scope ranging from advisory alone to embedded hands-on implementation and ownership. Over-engineering happens when the builder has only ever seen the technical side. I spent 15 years managing client relationships before I learned to code, so I see both the business problem and the system that solves it. The implementation I recommend comes from your call volume, your economics, and the systems you already run. Sometimes that means a managed platform configured well. Sometimes it means custom infrastructure you own. I will walk you through that decision, and then I will build the version your business calls for. I take a limited number of engagements at a time. If voice AI matters to your business and nobody owns the technical side, book a consultation. One hour, and you will leave knowing where your build stands and what to do next.