SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Glia is the #1 Banking AI platform serving 700+ banks and credit unions. Team Vector builds customer-facing voice AI agents that handle real conversations—from balance checks to loan inquiries—with industry-leading latency and safety guarantees.
In this role, you will design and build the core agentic voice AI framework, including conversation orchestration, tool calling, and runtime architecture. You'll develop the harness controls (guardrails, grounding, policy enforcement, verification layers) that make LLM agents trustworthy for banking's regulatory environment. You'll optimize latency across the entire real-time voice pipeline—speech recognition, LLM inference, tool calls, and synthesis—treating milliseconds as a first-class engineering constraint. You'll help unify runtime architecture to reduce configuration friction and enable rapid deployment of new banking use cases. You'll maintain production stability while shipping next-generation capabilities, and collaborate across distributed teams in Europe and Vancouver.
The tech stack includes Python, Elixir, TypeScript/React, PostgreSQL, DynamoDB, AWS, Kafka, Kubernetes, Docker, Terraform, DataDog, and AI tools like Gemini and Claude.
REQUIREMENTS:
- Production AI systems experience with deep understanding of LLM architecture: prompting, context management, tool use, agentic flows, retrieval, grounding, and evaluation. Real-time or voice AI experience strongly preferred; familiarity with agent SDKs (Claude Agent SDK, OpenAI Agents SDK) a plus.
- Strong architecture and technical judgment; ability to own complex component design, make sound tradeoffs, and identify risk early.
- Track record of fast, reliable delivery without cutting corners; experience verifying and standing behind agent-generated code.
- Active fluency with AI-assisted development tools (Claude, etc.); ability to direct agents, verify output, and know their strengths and limitations.
- Proactive problem-solving; ability to identify issues early, partner with PMs, and drive projects from concept to delivery.
- Adaptability in fast-paced environments with shifting priorities and evolving AI landscape.
- Strong learning velocity; curiosity and ability to quickly pick up new tools and practices.
- Clear written and verbal communication; proven success on distributed teams. Must reliably overlap at least 2 hours daily with Pacific Time (9:00–11:00 AM PT) for Vancouver collaboration.
- Leadership instincts; prior experience as tech lead, team lead, or mentor preferred.
