4 hours ago
Remote, EstoniaSenior
Responsibilities
- Design, build, and ship the core agentic voice AI framework, including conversation orchestration, tool calling, and runtime architecture.
- Develop guardrails, grounding, policy checks, evaluation, and other controls for trustworthy LLM agents in banking.
- Profile, measure, and optimize latency across the real-time voice pipeline.
- Unify runtime architecture and reduce configuration friction for new banking use cases.
- Maintain the stability and reliability of existing production systems.
- Collaborate with product, design, and engineering teams across Europe and Vancouver to deliver solutions.
- Use AI-assisted development tools and help establish standards for reviewing and verifying agent-generated code.
- Set direction, unblock teammates, identify risks, and raise engineering quality.
Requirements
- Production AI systems experience and strong understanding of LLM architecture, prompting, context management, tool use, agentic flows, retrieval, grounding, evaluation, and latency, cost, and quality tradeoffs.
- Experience owning the design of complex components and making sound technical tradeoffs.
- Track record of fast, reliable delivery, project de-risking, quality ownership, and verification of agent-generated code.
- Hands-on fluency with AI-assisted development tools such as Claude.
- Ability to proactively identify problems, partner with product managers, and drive projects from idea to delivery.
- Adaptability and curiosity in a rapidly changing technical environment.
- Clear written and verbal communication and ability to collaborate on a distributed team.
- Ability to reliably overlap at least two hours daily with Pacific Time from 9:00–11:00 AM PT.
- Prior technical leadership, team leadership, or mentorship experience is preferred.
- Real-time or voice AI experience and experience with agent SDKs such as the Claude Agent SDK or OpenAI Agents SDK are preferred.
Benefits
- Estonia-based role on a global team collaborating across Europe and Vancouver.
- Distributed-team schedule requiring at least two hours of daily overlap with Pacific Time.
- Opportunity to build customer-facing banking AI used by more than 700 banks and credit unions.
Tech Stack
Amazon DynamoDBApache KafkaArgo CDAWSDatadogDockerElixirJavaScriptJenkinsKubernetesPostgreSQLPythonReactTerraformTypeScript
Categories
About Glia
Glia builds a SaaS platform for banks and credit unions that unifies voice and digital customer interactions and adds banking-trained AI agents. Its products include a Banking AI Operating System plus co-browsing, chat, and voice tools that integrate with core systems and contact centers. Founded in 2012 and headquartered in New York, the privately held company sells to financial institutions in multiple markets.