Role overview

Own AI features end-to-end: translate product problems into AI solutions, prototype quickly, productionize, and continuously improve based on real usage and metrics.
Build and maintain production-grade AI services using Python and/or Node.js, integrated into RIVO’s backend architecture.
Develop the AI foundation for RIVO: shared building blocks, standards, templates, APIs, evaluation tooling, guardrails, and observability.
Design and implement prompting strategies, function/tool calling flows, and structured output patterns to achieve reliable results.
Build and optimize RAG pipelines (retrieval, ranking, chunking, context construction) and integrate with vector databases and hybrid retrieval techniques.
Implement evaluation and monitoring: offline testing, automated regression suites, online metrics, cost monitoring, and quality dashboards.
Work closely with Product, Engineering, and Domain Experts to ensure solutions are scalable, secure, and aligned with business value.
Research and adopt the best-fit models and approaches (OpenAI/Anthropic/open-source), including routing, fallback strategies, and cost/performance optimization.
Write clean, scalable, well-tested, and well-documented code.

What we're looking for

Knowledge and Experience**

3–6 years of backend development experience in Python and/or Node.js.
2+ years building AI/LLM-based solutions in production (not just prototypes). Demonstrated ability to take AI solutions from idea* production, including reliability, monitoring, iteration, and stakeholder alignment.
Strong understanding of LLM capabilities, limitations, and best practices (prompt design, tool/function calling, structured outputs, hallucination mitigation).
Experience integrating AI systems into real backend services (REST APIs, auth, async workflows, event-driven patterns).
Strong engineering fundamentals: system design, testing, versioning, scalability, maintainability.

Experience with AWS (Lambda/ECS/EKS, API Gateway, S3, etc.).
Hands-on experience with vector databases and embeddings (Pinecone, Weaviate, Qdrant, OpenSearch, pgvector).
Strong understanding of RAG and hybrid retrieval (BM25 + embeddings, reranking, filters, metadata strategies).
Experience with LLM evaluation frameworks and tooling (prompt/unit tests, golden sets, offline evals, A/B testing).
Familiarity with open-source LLMs and deployment patterns (vLLM, llama.cpp, model quantization).

Skills

Builder mindset: pragmatic, execution-focused, and comfortable working in ambiguity.
Strong ownership and accountability - you ship, you measure, you improve.
Excellent communication and collaboration skills.
Strong organizational and prioritization abilities in a fast-moving environment.
Proficient spoken and written English

Used for matching and alerts on DevFound

Ai Ai Engineer