Senior Agentic AI Engineer

ImagineX is a tech company that deploys AI-assisted teams to build and secure mission-critical enterprise solutions with our clients – spanning software, data, and AI. Structured like a software company, not a traditional consultancy, we blend deep technical expertise with authentic values, achieving world-class satisfaction (NPS 91). Our dedicated teams specialize in software, data, and AI across the U.S. and LATAM, bridging the gap between boutique agility and enterprise scale.

We are seeking a Senior AI Software Engineer with deep hands-on experience designing, building, and deploying Agentic AI systems in production. This is not an “AI-curious” or purely experimental role—we are looking for a practitioner who has shipped autonomous agentic architectures, understands the complex realities of productionizing LLM-driven workflows (reasoning loops, state management, tool-use orchestration, and guardrails), and can directly shape both technical and product direction.

1. Agentic System Architecture & Orchestration
   • Design, build, and deploy production-grade autonomous agent systems, multi-agent frameworks, and stateful orchestration pipelines in Python.
   • Implement tool-use reasoning loops, dynamic planning, function-calling patterns, and memory architectures (short-term, long-term, semantic memory).
   • Evaluate, benchmark, and integrate state-of-the-art foundation models (OpenAI, Anthropic, open-weight models) with a focus on latency, cost, determinism, and reasoning capabilities.
2. Enterprise Integration & Tool Capabilities
   • Connect autonomous agents securely to enterprise tools, databases, APIs, and external execution environments.
   • Build robust retrieval pipelines, hybrid search systems, and vector databases to supply real-time context to agentic workflows.
3. Productionization, Reliability & Observability
   • Implement evaluation frameworks (evals), guardrails, tracing, and observability to monitor agent behavior, handle edge cases, prevent hallucinations, and ensure deterministic outputs in production.
   • Design fallback mechanisms, human-in-the-loop (HITL) execution paths, and retry logic for high-reliability enterprise environments.
4. Technical Leadership & Client Strategy
   • Own the end-to-end technical direction for AI initiatives, translating complex business challenges into production-ready agentic architectures.
   • Partner with client leaders and internal engineering teams to shape product roadmaps for enterprise AI execution.
   • Lead rapid prototyping and continuous experimentation to validate novel agent capabilities before scaling.

• Core Python Engineering: Expert-level Python skills with a track record of building scalable, asynchronous, production-grade systems.
• Agentic AI & Orchestration: Direct experience building and productionizing agentic frameworks (e.g., LangChain/LangGraph, AutoGen, CrewAI, LlamaIndex, or custom in-house agent loops).
• LLM Tooling & Function Calling: Strong understanding of structured outputs, JSON mode, tool calling, multi-step reasoning, and context window management.
• Retrieval & Memory Architectures: Hands-on experience with vector search engines (Pinecone, Qdrant, Weaviate, pgvector), embeddings, and RAG/advanced memory retrieval patterns.
• Production Guardrails & Evals: Practical experience evaluating agent performance, tracking telemetry (e.g., LangSmith, Phoenix, Weights & Biases), and implementing safety guardrails.
• High-Autonomy Execution: Ability to navigate ambiguity, drive technical strategy, and deliver end-to-end features independently.

• Multi-Agent Collaboration: Experience building systems where specialized agents collaborate, delegate, or critique outputs hierarchically.
• Cloud-Native & Distributed Systems: Experience with event-driven architectures, serverless functions, and cloud platforms (Azure AI, AWS Bedrock, GCP Vertex AI).
• Enterprise Security: Understanding of RBAC, data privacy, and secure execution sandboxes for code-interpreter agents.