Lead AI Back-End Engineer

The impactyou’llmake as a Lead AI Back-End Engineer

•Set the technical architecturefor our agent powered platform, choosing the right mix of LLMs, retrieval technologies, agent frameworks, and microservice patterns to meet reliability, cost, quality, and latency targets.

•Drive architecturefor high throughput APIs built with .NET, C#, Python 3.11+,FastAPIasync,SQLModel, and Semantic Kernel, from design documents through production rollout.

•Lead multi agent orchestrationacross handoff, sequential, parallel, and supervisor patterns, combining knowledge grounded and tool calling agents acrossOpenAI GPT 5.6 Sol, Terra and Luna, Google Gemini 3.7 Flash, Anthropic Claude Sonnet 5 and Opus 5,xAIGrok 4.6, and future model providers.

• Guide teams implementingRetrieval Augmented Generationon Azure AI Search,pgvector, Chroma, and equivalent vector and hybrid retrieval platforms, ensuring index quality, filtering, evaluation, and safety controls.

• Ownend to end CI/CD pipelinesusing Bitbucket Pipelines, Jenkins, or similar platforms, including linting, type checking, security scanning, testing, containerization, and deployment.

•Mentor and hireengineers and adoptAI coding agents such as Cursor and Claude Codeas engineering force multipliers whilemaintainingstrong code quality, security, and review standards.

• Championobservability and FinOpsfor LLM workloads using structured JSON logging,OpenTelemetrytracing,Langfuse, evaluation frameworks, and cost dashboards, keeping latency and cost per request within defined SLOs.

• Partner with Product and Security to translate business goals, compliance requirements, emerging AI capabilities, and user feedback into a pragmatic technical roadmap.

Must have experience

• 7+ yearsbuilding and scaling production backend systems, including2+ yearsin a technical lead, staff, principal, or equivalent role.

• Expert in REST API design and development usingPythonFastAPIor .NET C# APIs, with strong experience in dependency injection, middleware, profiling, performance optimization, and distributed systems.

• Hands on leadership experience withSemantic Kernel or equivalent agent and LLM frameworks, including production agent orchestration, tool calling, structured outputs, andmulti stepworkflows.

• Delivered at least one productionRAG system or pipelineusing a vector or hybrid search platform such as Azure AI Search,pgvector, Chroma, or equivalent, with measurable latency and quality KPIs.

• Deep PostgreSQLexpertiseplusSQLModel,SQLAlchemy2, and Alembicmigrations at scale.

• Proventrack recordintegrating multiple frontier and cost optimized model providers, such asOpenAI GPT 5.6, Gemini 3.x, Claude 5, Grok 4.x, or equivalent, including model routing, structured output, reasoning, tool calling, and fallback strategies.

• Fluency withPoetry, Docker, GitHub Actions, Azure DevOps, Jenkins, or Argo, along with infrastructure as code fundamentals and blue green or canary release strategies.

• Strong people leadership through code reviews, architectural guidance, technical mentoring, roadmap planning, hiring, and cross team communication.

Nice to have

• Experience with message queue and event drivenarchitecturesusing RabbitMQ, Kafka, Azure Service Bus, or equivalent.

• Experience with GPU inference fleets,self hostedopen weight models, or serverless model hosting.

• Experience withmodel gateways, intelligent model routing, MCP, agent interoperability, prompt caching, and LLM cost optimization.

• Familiarity with automatic evaluation pipelines, agent evaluations, LLM as judge techniques, safety guardrails, andcost awareprompt and context engineering.

Why join theNewfoldAI team?

You’llsteer the core intelligence behind our AI products, building an LLM agnostic, agent driven platform that serves millions of users while meeting enterprise grade standards for reliability, security, governance, and economics. If you thrive on significant technical challenges, enjoy mentoring strong engineers, and want the opportunity to shape both architecture and engineering culture,we’dlove to meet you.