Role Summary

We are seeking a Senior AI/Integration Engineer to design, build, and operationalize LLM-powered applications, AI copilot experiences, and intelligent API orchestration layers. You will be at the center of our AI engineering practice — integrating foundation models into production systems, building agentic AI workflows, and creating the platform capabilities that enable the broader team to deliver AI-powered solutions at scale.

This is a hands-on, high-impact engineering role. You will work across the full stack of modern AI application development: from prompt engineering and model evaluation to API design, agent orchestration, and production deployment. You should be equally comfortable fine-tuning a retrieval pipeline as you are designing a resilient microservices architecture. 

Key Responsibilities

LLM Integration & Application Development

• Design and implement production-grade integrations with foundation models: OpenAI (GPT-4o, GPT-4o-mini, o3), Anthropic (Claude 3.5/4.x), Google (Gemini 3.x), Meta (Llama 3/4), and open-source models via vLLM, Ollama, or Hugging Face
• Build and optimize RAG (Retrieval-Augmented Generation) pipelines — including document ingestion, chunking strategies, embedding generation, vector storage, retrieval ranking, and response synthesis
• Implement advanced prompting techniques: chain-of-thought, few-shot, retrieval-augmented, tool-use, and structured output generation (JSON mode, function calling)
• Design model evaluation frameworks: automated benchmarking, A/B testing, human-in-the-loop evaluation, and regression testing for prompt changes
• Manage model selection, cost optimization, and latency tuning across multiple LLM providers

AI Copilot Development

• Design and build AI copilot experiences — context-aware assistants embedded in applications that augment user workflows with intelligent suggestions, automated actions, and natural language interfaces
• Implement conversational memory management: short-term (session), long-term (user profile), and shared (organizational knowledge)
• Build copilot features including intelligent document summarization, automated report generation, conversational data exploration, and guided workflow assistance
• Design copilot safety layers: content filtering, hallucination detection, confidence scoring, citation generation, and graceful degradation
• Instrument copilots with usage analytics, feedback loops, and continuous improvement mechanisms

Agentic AI & Orchestration

• Architect and build multi-agent AI systems using frameworks like LangGraph, CrewAI, AutoGen, Semantic Kernel, or custom orchestration layers
• Design agent tool-use interfaces: function calling, MCP (Model Context Protocol) servers, and API tool definitions with proper schema validation and error handling
• Implement agent planning and reasoning patterns: ReAct, Plan-and-Execute, Tree-of-Thought, and reflection/self-correction loops
• Build agent guardrails: execution sandboxing, approval workflows for high-stakes actions, resource limits, and audit logging
• Design agent-to-agent communication and coordination for complex multi-step workflows

API Orchestration & Platform Engineering

• Design and build API orchestration layers that compose multiple AI services, internal APIs, and external data sources into cohesive workflows
• Implement API gateway patterns for AI services: rate limiting, request routing, model fallback chains, caching, and usage metering
• Build event-driven architecture using message queues (Kafka, RabbitMQ, SQS) and workflow engines (Temporal, Prefect) for complex AI pipelines
• Design and maintain AI platform SDKs and internal libraries that abstract LLM provider complexity for other engineering teams
• Implement observability for AI systems: prompt/completion logging, token usage tracking, latency monitoring, and cost attribution

Integration Engineering

• Design and implement integrations with enterprise systems: CRM, ERP, ITSM, document management, and communication platforms
• Build robust API connectors with proper authentication (OAuth 2.0, API keys, mTLS), error handling, retry logic, and circuit breakers
• Implement data transformation and mapping layers between AI systems and downstream consumers
• Design webhook and event-driven integration patterns for real-time AI-powered workflows
• Ensure all integrations meet security, compliance, and data governance requirements