About this role
A Senior Software Engineer with a strong background in building production AI systems and an understanding of LLM architecture would find this role engaging. This position is central to the work of Team Vector, which is focused on developing the conversational AI that directly interacts with customers of banks and credit unions. The team's mission is to create and scale a fully agentic voice AI framework designed to handle high-value customer interactions efficiently. They aim to unify the runtime architecture, reduce setup friction, and maintain stability within existing systems.
This role involves deep technical work on the core framework, focusing heavily on the real-time performance constraints of voice AI, such as minimizing latency across speech recognition, LLM inference, and tool calls. A key challenge is developing robust guardrails, grounding mechanisms, and policy enforcement layers to ensure the agents are safe, accurate, and compliant within a highly regulated banking environment. The impact of this work is significant, as the agents built will handle live conversations for numerous financial institutions.
This role is ideal for an engineer who thrives on solving complex applied AI problems under strict operational constraints. Individuals who excel at system design, making critical tradeoffs between speed and quality, and who prioritize shipping reliable, high-quality outcomes would perform well in this environment.
What you will do
- Design, build, and deploy the core components of the agentic voice AI framework, encompassing conversation orchestration and runtime architecture.
- Develop and refine the control mechanisms, including guardrails, grounding, policy checks, and evaluation layers, to ensure the trustworthiness of the LLM agents.
- Optimize the real-time voice pipeline by profiling, measuring, and reducing latency across all stages, such as speech recognition and LLM inference.
- Work to consolidate the runtime architecture to eliminate configuration hurdles, enabling faster deployment of new banking use cases.
- Maintain the stability and reliability of existing production systems while developing future AI capabilities.
- Collaborate with product, design, and engineering teams across different regions to translate ambiguous requirements into functional outputs.
- Utilize AI-assisted development tools daily to establish standards for reviewing and verifying agent-generated code.
What you need
- Demonstrated experience building production AI systems is required, along with a solid understanding of LLM architecture, including prompting, context management, tool use, and agentic flows.
- Experience with retrieval and grounding techniques, as well as an understanding of the trade-offs between latency, cost, and output quality is essential.
- Strong architectural judgment is necessary to own the design of complex components and proactively identify potential risks.
- A track record of delivering fast and reliable results, including de-risking projects and ensuring quality assurance, is required.
- Experience with real-time or voice AI systems is considered a strong advantage, as is familiarity with agent SDKs like the Claude Agent SDK or OpenAI Agents SDK.
- Proficiency in Python is a core coding language for this role.
- Experience in system design where AI contributes to code generation is valued.
Nice to have
- Experience with agent SDKs (such as the Claude Agent SDK or OpenAI Agents SDK)
Benefits
- A global team across Europe and Vancouver, Canada
Skills and tech stack
Questions candidates ask
What is the remote policy?
The team Vector works across Europe and Vancouver, Canada, indicating a globally distributed setup. Candidates will collaborate closely with teams on both continents.
What is the compensation structure?
The posting does not specify salary details for this role.
What is the primary technology stack?
The tech stack includes Python, Elixir, TypeScript, React (JavaScript), AWS services like RDS for PostgreSQL, DynamoDB, and S3, along with tools like Terraform, Kafka, and Kubernetes.
How is the hiring process structured?
The posting focuses on the technical requirements and desired experience level rather than detailing a specific hiring process.