Sole is in direct contact with the company and can answer any questions you may have. Email
Cloaked is a privacy-first identity company building consumer products that protect people's personal information at scale. We are an AI-native team that ships fast and operates lean. We are looking for a Voice Agent Engineer to design, build, and harden the AI voice agents that interact directly with our users and the systems that protect them.
This is a hands-on contract engagement, not a delineated-task role. You will own voice-agent features end to end — from real-time conversation design through telephony integration and production reliability — in a privacy-sensitive, high-compliance environment. Because this role touches sensitive, regulated systems, it is open only to US citizens based in the US.
Voice Agent Development
● Build and iterate on LLM-driven voice agents that hold natural, low-latency conversations with users.
● Integrate speech-to-text, text-to-speech, and turn-taking logic into reliable real-time pipelines.
● Tune prompts, tools, and guardrails so agents stay accurate, on-task, and safe under real-world conditions.
Telephony & Real-Time Systems
● Wire voice agents into telephony and streaming-audio infrastructure with attention to latency, jitter, and failure modes.
● Handle call state, interruptions, and fallbacks gracefully so the experience degrades safely, never silently.
Privacy & Reliability
● Operate within strict handling requirements for sensitive and regulated data.
● Instrument agents for observability — measure quality, latency, and failure rates, and act on them.
AI-Native Execution
● Use modern AI coding tools fluently to ship at high velocity across multiple workstreams.
● Run your own productivity harness — managing parallel agent streams and many concurrent tickets is the default, not the exception.
● Strong production experience building voice, conversational AI, or real-time audio systems.
● Fluency with LLM application development — prompting, tool use, evaluation, and guardrails.
● Comfort with latency-sensitive, real-time backend systems and their failure modes.
● US citizenship and US residency (non-negotiable for this role, which touches sensitive, regulated systems).
● Proven ability to hit the ground running with minimal onboarding and operate autonomously in ambiguity.
● Daily fluency with AI-assisted development tools and a personal setup for working quickly across parallel tasks.
● Exceptional written and verbal English communication.
● Experience in privacy, identity, security, fintech, or other high-compliance domains.
● Familiarity with major telephony / voice platforms (e.g., Twilio, LiveKit, or similar).
● Experience evaluating and red-teaming conversational agents.
● Startup background where you owned features end to end.
● Contract engagement, time-and-materials; begins with a two-week trial.
● High autonomy, high accountability — top-percentile execution expected.
● AI-native by default; ramps quickly and flexes across the stack as the product requires.
● East Coast time-zone overlap with the team.