🚀 About Murphy AI

Murphy AI is revolutionizing debt collection through conversational artificial intelligence. We're building an advanced debt recovery platform where AI agents conduct natural, empathetic conversations across multiple channels—with a special focus on voice interactions at scale.

Our AI agents don't just send emails or SMS. They pick up the phone, understand human speech, respond intelligently, and negotiate—all while maintaining the nuance and empathy that complex financial conversations require. We're talking about real-time speech processing, emotional intelligence, and conversational flows that adapt on the fly.

The challenge? Making our agents sound so natural that debtors forget they're talking to AI, while handling the complexity of financial negotiations, regulatory compliance, and emotional conversations—and making the system increasingly more intelligent with every interaction.

🎯 The Role

As our Senior AI Engineer, you'll be at the forefront of our voice and conversational AI capabilities. You won't be building generic chatbots—you'll be architecting the intelligence behind thousands of daily phone calls and multi-channel conversations that recover millions in debt while treating people with dignity.

You'll work on challenging problems where generative AI meets real-world voice interactions: designing conversational flows that handle complex negotiations, optimizing speech systems for natural dialogue, and adapting our platform to work seamlessly for different clients with varying requirements and constraints.

🔥 What Makes This Role Unique

• Real-world voice AI impact: Your contributions directly power live phone conversations with real people
• End-to-end ownership: From conversation design to deployment and optimization
• Complex conversational challenges: Navigate payment negotiations, objection handling, and emotional conversations
• Cutting-edge tech stack: Latest LLMs, advanced TTS/STT systems, real-time processing
• Immediate feedback loops: See your AI agents in action and iterate rapidly

🛠️ What You'll Build

🎙️ Advanced Conversational Systems

• Design and optimize real-time voice interaction pipelines for natural, low-latency conversations
• Build context-aware conversational flows that adapt tone, pace, and strategy based on conversation state
• Implement quality assurance systems that monitor and improve conversation performance

🧠 Conversational Intelligence

• Develop AI agents that understand context, emotions, and intent in real-time conversations
• Build intelligent conversation routing that knows when to escalate, when to negotiate, and when to empathize
• Design prompt strategies and agent behaviors that maximize both recovery rates and user experience
• Experiment with different approaches to improve conversation outcomes continuously

⚡ Production AI Infrastructure

• Optimize latency-critical pipelines for natural conversation flow
• Build scalable orchestration systems handling hundreds of concurrent interactions
• Design A/B testing frameworks for voice and text interactions
• Ensure reliability, monitoring, and observability of AI systems in production

🔧 AI Tooling & Innovation

• Create internal tools for conversation analysis, quality evaluation, and performance monitoring
• Build simulation and testing frameworks for conversational edge cases
• Research and experiment with new AI technologies to keep Murphy ahead of the curve
• Explore innovative approaches to voice synthesis, speech recognition, and conversational design

🎪 The Technical Playground

AI & Voice Stack:

• LLMs: OpenAI, Anthropic and others
• TTS: ElevenLabs (experience with these or similar tools is a plus)
• STT: Whisper, AssemblyAI (experience with these or similar tools is a plus)
• Real-time Communication: WebSockets, Livekit (experience with these or similar tools is a plus)
• AI Observability: LangFuse

Infrastructure:

• Backend: Node.js microservices, TypeScript, Python
• Messaging: Kafka for inter-service communication
• Monitoring & Observability: LangFuse, Datadog
• Data: PostgreSQL, vector databases

🏆 What We're Looking For

🎯 Core Requirements

• 4+ years of software development experience with a solid engineering foundation
• Proven experience deploying Gen AI solutions in production (LLMs, RAG pipelines, AI agents)
• Strong interest and passion for voice AI and TTS/STT technologies - you're excited about conversational AI and want to dive deep into this area
• Fullstack capabilities: You understand both backend systems and how they integrate with user-facing applications
• Collaborative mindset: Experience mentoring teammates and contributing to team growth

🌟 Nice to Have (Desirable)

• Background in speech processing, audio preprocessing, or acoustic systems
• Experience with conversational AI frameworks and voice flow design
• Knowledge of voice quality metrics and evaluation methodologies
• Understanding of WebSocket protocols and real-time communication systems
• Experience with supervised learning, model training pipelines, or classical ML approaches