AI Platform & Integration Engineer

AI Platform & Integration Engineer We are building an enterprise intelligence layer that connects data, AI models, business systems, reasoning, workflows, and user experiences. We are seeking an AI Platform & Integration Engineer to build the infrastructure behind intelligent enterprise applications. This person will focus less on visual frontend development and more heavily on AI architecture, model orchestration, APIs, data connectivity, enterprise integrations, reliability, security, and production AI systems. What You’ll Own Design and implement multi-model AI infrastructure Integrate and manage model access through OpenRouter and direct model-provider APIs Build intelligent model routing based on task, latency, cost, privacy, reliability, and capability Implement model fallbacks and resiliency strategies Build AI tool/function-calling systems Develop agentic and multi-step workflow architectures Connect AI systems securely to enterprise databases, APIs, SaaS platforms, CMS environments, search tools, CRMs, and proprietary systems Build RAG and enterprise knowledge-retrieval systems Implement semantic search and entity resolution Design contextual and long-term application memory Build enterprise data pipelines and ingestion services Normalize and reconcile fragmented data Develop secure data-access layers for AI Implement permissions and policy enforcement outside the model Establish observability for AI requests, costs, latency, tool calls, failures, and model performance Build evaluation frameworks for accuracy and reliability Optimize inference cost and performance Support private, hybrid, cloud-agnostic, or client-specific deployments Collaborate with Solution Architects to turn future-state designs into working infrastructure Support Full-Stack Engineers consuming the AI and data services you create AI Experience We Want Deep experience with several of the following: OpenRouter OpenAI Anthropic Google Gemini Open-source models Multi-model orchestration Tool/function calling Structured outputs Agent architectures RAG Embeddings Semantic retrieval Vector databases Model routing Model evaluation Guardrails Context management Prompt engineering AI observability Inference optimization Experience with technologies such as: LangGraph LlamaIndex LangChain LiteLLM MCP / Model Context Protocol Vercel AI SDK pgvector Pinecone Weaviate Qdrant is valuable. We are not looking for someone who relies exclusively on one framework or one model provider. Data & Integration Experience Strong experience with: Python TypeScript / Node.js PostgreSQL SQL Redis REST APIs Webhooks Authentication OAuth API gateways Data pipelines ETL/ELT Data normalization Entity resolution Data quality Event-driven systems You should be comfortable being handed multiple databases and APIs that were never designed to work together and developing a clean, secure integration architecture. Architecture Philosophy We expect you to understand that an LLM should not control everything. You should know how to determine when to use: Deterministic business rules Traditional algorithms Search Database queries Machine learning Knowledge graphs Vector retrieval LLM reasoning External tools Human approval Our goal is reliable enterprise intelligence—not AI theater. Security & Enterprise Readiness You should understand: Data minimization Role- and attribute-based access Tenant isolation Encryption Secrets management Auditability PII handling Model-provider data policies Prompt injection risks Tool permissions Least-privilege access Production monitoring Failure recovery How We Work This is a high-accountability environment. We expect: Proactive communication Dependability Ownership Clear documentation Early escalation of blockers Thoughtful technical pushback Respect for scheduled meetings and commitments Independence without isolation If something changes, communicate. If you see a problem, raise it. If you own something, drive it. Ideal Candidate You likely have: 5+ years of software, backend, platform, ML, AI or data-engineering experience Significant hands-on experience with production generative AI systems Experience integrating complex enterprise environments Strong Python skills Strong API and data architecture experience Experience with multiple model providers Production deployment experience Excellent troubleshooting skills The ability to explain technical decisions to nontechnical stakeholders Engagement: Contract, with potential to expand based on performance and company needs. Candidates should be prepared to demonstrate previous systems they have personally built and may participate in a live technical architecture exercise.