AI Agent Engineer (Python) / Full-Stack experience

About Orbio

Orbio is an AI-native human capital management platform building a Digital Workforce for HR.

We design and deploy agentic systems that take ownership of operational execution across hiring, onboarding, and employee operations — while keeping humans in control.

We work with large, complex organizations across staffing, retail, hospitality, and healthcare. Our customers rely on Orbio to reduce operational friction, improve speed and quality, and scale their teams without proportional headcount growth.

Orbio is not a system of record. We sit on top of existing HR systems and own the execution and coordination layer — where day-to-day work actually happens.

Position Overview

Orbio is looking for an AI Agent Engineer to build, operate, and continuously improve the reliability of our agentic AI systems in production.

In this role, you will focus on how agents behave, decide, fail, recover, and scale in real enterprise environments. Your work will ensure that LLM-powered agents are observable, testable, cost-efficient, and trustworthy when handling real operational workflows.

You will help build an end-to-end, production-grade B2B platform with a strong focus on LLMs, autonomous agents, and conversational AI, contributing to the technical foundations that scale across customers and use cases.

All the desired Technical Skills we are looking for

Agent Testing & Evaluation

• Building agent evaluation pipelines (consistency, accuracy, reliability)
• Designing test harnesses for non-deterministic systems
• Regression testing for prompt and model changes
• Synthetic data generation for agent testing
• Performance benchmarking and latency profiling

Agent Monitoring & Operations at Scale

• Observability for agent systems (tracing, logging, metrics)
• Cost tracking and optimization across model calls
• Alerting on agent failures, hallucinations, and degraded performance
• Debugging complex multi-step agent behaviors in production
• Guardrails, safety layers, and graceful degradation strategies

LLM Expertise

• Prompt engineering, chaining, and caching strategies
• RAG pipelines and vector databases
• Model selection, fine-tuning, and cost optimization
• Understanding of evaluation metrics and benchmarks

Python & Backend

• Async programming, Django/Channels, real-time ASGI
• Production-grade application development
• Message queues (Celery + Redis) and high-throughput systems
• Testing, logging, and performance optimization

Conversational AI

• Voice pipelines: STT, TTS, turn-taking, interruption handling
• Telephony integration (Twilio) with latency optimization
• Conversation flow design and fallback strategies

Full-Stack & DevOps

• Frontend experience (React preferred)
• CI/CD pipelines and cloud deployment (AWS/GCP)
• Deploying and operating AI systems in production

What Success Looks Like at Orbio

• Agents behave consistently and predictably across customers and workflows
• Failures, edge cases, and degraded behavior are detectable and recoverable
• Agent performance, latency, and cost are measurable and actively optimized
• New agent capabilities ship safely without breaking existing behavior
• Customer trust in autonomous workflows increases over time

Culture at Orbio

Orbio is a high-bar, high-intensity environment. We optimize for impact, speed, and ownership — not comfort.

We operate as one team, with a flat structure and extreme ownership. We ship fast, learn from production, and challenge each other directly because we care about building systems that hold up under real-world complexity. Titles matter less than outcomes. Customers come first. Momentum matters.