Senior AI Engineer (Agentic Workflows)

TL;DR

Senior AI Engineer (Agentic Workflows): Building and shipping personalized, self-improving agentic workflows deployed at scale with an accent on runtime design, memory systems, evaluation harnesses, and production infrastructure. Focus on designing stateful agent systems, retrieval pipelines, voice experiences, and reliability patterns for trustworthy production agents.

Location: Remote; overlap with Americas timezones for collaboration; reliable high-speed internet

Company

Supernal helps small-to-medium businesses hire their first AI employee through intelligent, agentic workflows on a proprietary platform that deliver value-generating AI teammates for real business processes.

What you will do

• Ship end-to-end user-facing agent experiences from prototype to production with iteration based on usage.
• Architect and implement stateful agent systems including workflows, tool calling, memory, retrieval, and human-in-the-loop.
• Build voice features end-to-end such as realtime speech agents, voice UI/UX, prompt/audio routing, and guardrails.
• Develop and own evaluation harness with test sets, automated scoring, tracking, and rollout patterns.
• Design data and retrieval pipelines with chunking, hybrid retrieval, re-ranking, caching, and multi-tenant isolation.
• Integrate with platform services like Django/DRF, async queues, PostgreSQL/pgvector, Kubernetes, and establish observability, reliability, and security.

Requirements

• Overlap with Americas timezones for collaboration; reliable high-speed internet.
• Strong software engineering fundamentals in design, testing, code quality, performance, security.
• Production experience deploying AI systems in front of users.
• Experience building agentic or LLM-powered systems with memory and tool use.
• Comfort across application and infrastructure layers: APIs, jobs, data stores, deployment.
• Hands-on with agent frameworks like LangChain/LangGraph, LlamaIndex, or equivalent; strong retrieval/vector search knowledge.

Nice to have

• Vector databases/search stacks like Pinecone, Chroma, Weaviate, Qdrant, pgvector.
• Designing evaluation systems including offline eval, monitoring, regression.
• Voice/real-time systems with streaming, WebRTC, STT/TTS integration.
• Durable workflows with Temporal or similar.
• Observability tooling like OpenTelemetry, Datadog.
• Shipping multi-tenant SaaS with privacy boundaries.

Culture & Benefits

• Full-time remote role with competitive salary for senior level.
• Partner closely with platform engineering on core services.
• Focus on engineering rigor, observability, reliability, and security for AI agents.

Hiring process

• Interview focus: system design for agentic apps, retrieval/RAG design, production practices, tradeoffs under uncertainty.