Senior AI Engineer

Senior AI Engineer (OpenAI, Lang Chain/Lang Graph, Type Script, Supabase)

About Boldspace&Bold Stream™

Boldspaceis a6-year-oldfull-service

Brand& communicationsagencythat hasexperienced year-over-year growth sinceitsfounding.

Our seasoned agency-side co-founders felt there was acriticallack of industry transparencyindata and measuring the performance ofclient work,so welaunched

BoldStreamon day 1, our flagship data & analytics platform that helps brands measure and act on the impact of their communications in real time.

Wewill now continue to investheavily in AI-native capabilities: agentic workflows, retrieval-augmented insights, and intelligent automation across our client reporting and analytics stack.

We’relooking for a Senior AI Engineer to own end-to-end delivery of AI-powered features across

BoldLoopand related internal products.

You’llsit at the intersection of OpenAI’s latest agentic tooling,Lang Chain/Lang Graph, Type Script, and Supabase, building production-grade systems rather than prototypes.

This role is hands‑on and full-stack: from designing LLM‑powered workflows and data models to shipping polished front‑end experiences and observability for everything you release.

• Design and implement agentic workflows using OpenAI (Responses API, Agent Kit & Agent Builder) and Lang Chain/Lang Graph for use cases like analytics copilots, automated reporting, insight generation and internal tools.
• Build and tune prompts, tools and orchestration logic to make agents reliable: handle tool‑calling, multi‑step reasoning, fallbacks, and guardrails.
• Architect and implement RAG pipelines (embeddings, vector stores, retrieval strategies, caching) using Supabase and other data sources to ground models in Bold Loop and client data.

• Backend:
  Type Script/Node or edge functions, Supabase (Postgres + vector store), API design, schema design, migrations and performance tuning.
• Frontend:
  Modern Type Script front-ends (React / Next.js or similar) for analytics dashboards, workflows and internal tooling.
• Integrate OpenAI and other LLM providers in a way that is robust, observable, and cost‑aware (tracing, token usage monitoring, rate limiting, retries).
• Work closely with data engineers / analysts to expose Bold Loop metrics, models and datasets through AI features that feel intuitive to non‑technical users.

• Champion testing, observability and reliability for AI features: unit tests, integration tests, synthetic evals for prompts, and health checks for agent workflows.
• Contribute to and improve CI/CD pipelines, deployment strategies and cloud infrastructure supporting AI workloads.
• Use Git Hub, Linear and AI‑enhanced tooling (Cursor, Lovable, etc.) to move quickly while maintaining high standards.

• Translate business and product requirements into technical designs and implementation plans for AI features.
• Provide technical leadership within the AI /Bold Loop squad, mentoring engineers on LLM patterns, agentic architectures and best practices.

Partner with Product, Data, Client Services and Strategy to define AI roadmaps that unlock measurable value for clients.

• 5+years experience as a Full‑Stack / Backend Engineer, with at least 2+years building LLM‑powered products in production.
• Strong Type Script and JavaScript fundamentals; production experience with Node.js backends and modern front‑end frameworks.
• Hands‑on experience with Lang Chain and/or Lang Graph, ideally including multi‑agent or complex tool‑using workflows.
• Proven experience integrating OpenAI APIs (chat, Responses, Agents SDK) into production systems.
• Solid experience with Supabase or similar (Postgres, auth, edge functions, vector search) and data modelling for analytics and content.
• Deep understanding of RAG, prompt engineering, evaluation, and safety/guardrail patterns for LLM applications.
• Comfortable with CI/CD, cloud infrastructure and observability for distributed systems.

• Experience with additional AI frameworks (CrewAI, Lang Flow, semantic search stacks, etc.).
• Exposure to analytics or martech platforms,…