About the Role

We're seeking an experienced backend engineer to own and scale our multi-channel automated outreach platform. This system orchestrates thousands of voice, SMS, and email communications across lending workflows, with sophisticated scheduling, analytics, and quality monitoring capabilities.

What You'll Own

Voice Agent Infrastructure

• Telephony integration with dynamic agent configuration and worker dispatch
• Voice call analytics pipeline (evaluation, custom metrics, sentiment analysis)
• Real-time call quality monitoring (hallucination detection, redundancy analysis, pronunciation scoring)
• PDF reporting system generating comprehensive voice performance reports
• Call queue management and retry logic

Analytics & Monitoring Dashboard

• Interactive Flask-based debugging dashboard for real-time outreach tracking
• Multi-dimensional analytics (channel combinations, agent type breakdowns, spam detection)
• Custom evaluation metrics aggregation across voice, email, and SMS
• Agent performance visualization showing scheduled vs. actual communications with delay indicators

Database & Infrastructure

• PostgreSQL schema management with Alembic migrations.
• AWS ECS deployment pipelines (Bitbucket/Github Pipelines with OIDC authentication)
• Docker containerization and local development environment
• ECR image management and task definition updates

Technical Stack You'll Work With

Backend: Python 3, Flask, async/await patterns
Database: PostgreSQL, complex SQL queries
APIs: LiveKit SDK, RESTful endpoints, webhook handlers
Cloud: AWS (Cloud Watch)
CI/CD: Github Pipelines, Docker
External Services: Retell AI, various external APIs, Lendflow application API
Frontend: Vanilla JavaScript, CSS (for dashboard), HTML templates

Key Systems You'll Maintain

1. Voice Analytics Pipeline
2. Integration with external vendor for call quality evaluation
3. Custom evaluation metrics tracking with success rate calculations
4. Hallucination, redundancy, and pronunciation monitoring
5. Agent speak percentage and latency tracking2.Reinforcement Learning Data Collection
6. Vector store integration for similar example retrieval
7. Positive/negative example management for SMS/email responses
8. Similarity search with configurable thresholds and filtering
9. API Queue System
10. Asynchronous API request queuing with retry logic
11. FIFO processing with configurable batch limits
12. Failed request tracking and replay capability

Scale You'll Handle

• Thousands of active workflows across multiple agent types
• Multi-channel campaigns (email + voice + SMS combinations)
• Real-time analytics across 30+ day windows
• Concurrent call processing and quality evaluation
• Complex database queries joining analytics and operations schemas

What Success Looks Like

• 99.9% outreach delivery reliability across all channels
• Sub-second API response times for analytics queries
• Zero data loss during migrations and deployments
• Comprehensive monitoring catching issues before they impact customers
• Self-service debugging tools reducing operational overhead

Must-Have Experience

• 3+ years building production Python REST APIs using Flask, Django, FastAPI, or comparable frameworks
• Strong relational database skills with PostgreSQL preferred (complex queries, schema design, migrations, indexing, performance tuning)
• Container orchestration experience with AWS ECS/Fargate, Kubernetes, or similar platforms
• Docker containerization and CI/CD pipeline management
• RESTful API design and webhook integration
• Experience with LLMs, NLP frameworks, and prompt engineering techniques
• Experience with Git and version control systems

Nice-to-Have

• LiveKit or similar WebRTC platform experience
• Experience with telephony or real-time communication systems
• Voice AI/LLM agent platform integration (Retell, Vapi, etc.)
• Alembic or other database migration tools
• PDF generation with ReportLab
• Vector database or embedding-based search
• Github Pipelines or similar CI/CD

You'll Work On

• Scaling the voice agent system to handle 10x call volume
• Building new analytics dashboards for campaign performance
• Improving outreach timing algorithms for better engagement
• Integrating new communication channels or AI evaluation providers
• Optimizing database queries for faster dashboard load times
• Creating comprehensive alerting for system health

This role is perfect for someone who:

• Loves building systems that "just work" at scale
• Gets excited about AI and automation
• Desires an environment where your code directly impacts business outcomes
• Wants to see their work handling thousands of real customer interactions daily
• Can multitask and prioritize