At Veracity, we aim to be a different kind of insurance partner – one that is free from outside investors, venture capital, or the pressures of a corporate parent.
Ours is a culture of empowerment – one that believes in effort, results, and accountability. We believe that transparency fosters trust, trust fosters growth, and that growth drives innovation. Our commitment to rigorous evaluation and relentless execution lead to rapid evolution.
We answer only to the small business owners we serve, and this independence allows us to stay focused on what matters most: helping their businesses thrive by providing expert guidance and best-in-class insurance policies.
We’re growing fast and want you to be a part of it!
We're seeking a technically sharp and organizationally influential Senior Applied AI Engineer to join our Engineering team. Veracity is building a new multi-tenant, API/MCP-first insurance platform where AI agents operate as first-class principals alongside human users – built on a deterministic core with an agentic layer. Money movement, binding, and compliance run on deterministic, auditable systems; agents handle orchestration, intake, document understanding, and workflow acceleration on top.
This is not a chatbot-building role. As a Senior Applied AI Engineer, you will do two jobs in one: build and harden production agent systems for a regulated insurance domain, and help define and implement our AI-first engineering and product principles – how the whole organization designs, builds, ships, and operates software with AI in the loop. This is a senior individual-contributor role with high architectural and organizational influence and a clear growth path toward technical leadership as the AI team scales.
Key Responsibilities
- Design and ship production agentic systems – multi-agent orchestration, tool use, RAG pipelines, and MCP-based integrations against platform APIs
- Build the reliability layer around agents – evals, guardrails, observability, cost controls, and regression testing – so agent behavior is measurable and defensible
- Implement maker-checker patterns so AI accelerates regulated workflows without ever autonomously touching money, bind, or compliance decisions
- Integrate agent capabilities with the deterministic core including canonical data model, Config Hub, and Control Tower task orchestration
- Co-author and implement our AI-first engineering principles – agent harnesses, AI-assisted code review, shared skills and prompt libraries, eval-gated CI, and dev-environment standards – then drive adoption across squads in Vilnius and the US
- Shape AI-first product principles – where AI belongs in the product and where it doesn't, human-in-the-loop patterns, agent UX conventions, and how AI capability estimates change scoping and build-vs-buy decisions
- Set the standard through your own delivery – demonstrating what AI-first engineering makes possible and establishing the reference patterns other engineers adopt
- Work directly with engineering leadership on the AI architecture roadmap and year-end production proof points
- Required to perform other duties as requested, directed, or assigned
Requirements and Qualifications
- 5+ years of software engineering experience with strong backend or full-stack foundations – APIs, event-driven systems, and data modeling
- Real production experience with LLM-powered systems – agent orchestration, tool calling, RAG, and structured outputs – not just prototypes or demos
- Demonstrated ability to make agentic systems reliable – evals, fallbacks, deterministic checkpoints, and cost and latency management
- Strong product judgment – you scope AI where it earns its place and default to deterministic code where it doesn't
- Evidence you've changed how others build, not just how you build – workflows, harnesses, or practices you introduced that a team adopted and kept using
- Clear written communication with comfort operating in an async-first