Ai Rag Jobs in Miami, FL (NOW HIRING)

The Lead AI Software Engineer will set the technical bar for design, build, and ship AI-native software. This is a hands-on, deeply technical role for an engineer who lives at the intersection of cloud platform architecture, generative and agentic AI, and AI-accelerated software delivery.
This is a future-facing role. The toolchain, models, and frameworks named below will evolve - we are hiring for the judgment, depth, and adaptability to evolve with them and to lead others through that change.
________________________________________
What You'll Do
Architect AI-native systems. Design secure, highly available, scalable applications and platforms on Google Cloud - from reference architecture through production. Own the hard decisions around availability targets, failure modes, data flow, latency, cost, and blast-radius containment.
Drive spec-driven development. Establish and champion an AI spec-driven development framework (e.g., BMAD Method, OpenSpec, GitHub Spec Kit) as the team's delivery discipline - translating intent into executable specifications that humans and AI agents build against, review against, and verify against.
Build with agentic AI. Design and deliver agentic systems: multi-agent orchestration, tool use, and retrieval - using the Agent2Agent (A2A) protocol for agent-to-agent interoperability and the Model Context Protocol (MCP) for tool and context integration, routed and governed through an agent / MCP gateway. Treat agents as first-class production software, with the same rigor for security, observability, and reliability as any other critical system.
Move fast on research and POCs. Use AI coding agents (Claude Code, Codex, Antigravity, and successors) to compress the cycle from idea to working prototype to validated POC. Run structured experiments, evaluate models and approaches, and bring back evidence, not opinions.
Engineer for the model layer. Apply Gemini and Vertex AI deeply - prompt and context engineering, grounding/RAG, tool calling, function/agent design, fine-tuning where warranted, and model selection trade-offs across quality, cost, and latency.
Make it secure by design. Bake security and privacy controls into the architecture from day one, not as an afterthought, including agent identity and access management (verifiable, least-privilege, fully auditable identities and credentials for AI agents and other non-human workloads), data loss prevention, prompt-injection and jailbreak defenses, and content/safety filtering.
Own quality and observability. Stand up evaluation harnesses, regression suites, and AI observability (quality/drift monitoring, tracing, FinOps/cost visibility) so that AI behavior is measurable, traceable, and accountable in production.
Lead technically. Be part of core AI foundation team to set engineering standards, review designs and code (human- and AI-generated), mentor engineers on AI-native practices, and raise the team's collective ceiling. Partner with architecture, security, platform, and product stakeholders to land outcomes.
 Estimated salary range for this position is $126148.63 - $163993.22 / year depending on experience.

Degrees:

• Bachelors.

Additional Qualifications:

• Bachelor's degree or higher in Computer Science or equivalent is required.
• 10 years of professional software engineering experience, with a strong track record of shipping production systems (not just prototypes).
• Deep, hands-on Google Cloud (GCP) expertise - compute, networking, IAM, data, and the AI/ML stack. Able to architect a secure, high-availability system on GCP and defend the design.
•  Domain depth in Gemini and Vertex AI - building real applications on the platform, including grounding/RAG, tool/function calling, and agent development.
• Extensive hands-on experience with AI coding agents - Claude Code, Codex, Antigravity, or equivalent - used for serious development, research, and rapid POC work (not casual autocomplete). You can speak to how you structure work for agents and where they help vs. hurt.
• Proven system-architecture ability - designing for security, high availability, scalability, fault tolerance, and cost-efficiency. Comfortable with distributed systems fundamentals and trade-off analysis.
• Hands-on Agent Gateway / MCP gateway experience - building, deploying, or operating an agent or Model Context Protocol (MCP) gateway as the governed control point for agent and tool traffic, including agents that interoperate over the Agent2Agent (A2A) protocol and tools/context exposed via MCP. This includes centralized routing, authentication and authorization, agent/tool registration and discovery, rate limiting and quotas, policy enforcement, and observability across agent-to-tool and agent-to-agent calls (e.g., Apigee X or equivalent gateway).
• Spec-driven / AI-assisted delivery experience - you have used (or stood up) a structured framework for building software with AI agents and can articulate why specifications-as-source-of-truth matters.
• Strong software-engineering fundamentals - at least one modern language at expert level (e.g., Python, Go, TypeScript/Node, Java), API design, testing, CI/CD, and Git-based workflows.
• Security-first mindset - secure SDLC, secrets management, least-privilege design, agent/non-human identity and access management, and awareness of AI-specific threats (prompt injection, data exfiltration, model abuse).
• Excellent communication - able to explain complex architecture to engineers and executives, write clear specs and design docs, and influence without authority.
  ________________________________________
  Preferred / Bonus Qualifications
      Depth at the gateway layer - Apigee X (or equivalent) experience using native proxy generation, MCP/agent gateway patterns, and AI-traffic governance at enterprise scale.
      Agent-platform & tooling - building standardized tools, skills and integration layers, agent registries, and reusable agent frameworks for other teams to consume.
      AI safety & governance tooling - DLP, AI safety/guardrail filters (e.g., Model Armor), and responsible-AI practices.
      AI observability & FinOps - drift/quality monitoring, evaluation pipelines, OpenTelemetry, and cost governance for LLM workloads.
      Regulated / compliance-heavy domain experience - building and operating systems in environments with strict data-protection, audit, and governance requirements.
      Scaled delivery context - SAFe or other scaled-agile environments; partnering with enterprise architecture and capital planning.
      GitLab / GitHub CI-CD at scale, and IaC (Terraform).
      Relevant certifications - Google Cloud Professional (Cloud Architect, ML Engineer, or equivalent).
      Open-source contributions, research, or public artifacts in AI engineering, agents, or developer tooling.

Minimum Required Experience: 10 Years