AI Agentic Platform Engineer

ROLE PROFILE & TECHNICAL STANDARD
AI Agentic Platform Engineer
Bridging cloud-native platform infrastructure, distributed systems, and cognitive autonomous agent
architectures.
1. Core Technical Requirements
Agent Orchestration & Standards
Platform & Compute Infrastructure
Framework Mastery: Hands-on expertise with
LangChain, LangGraph, AutoGen, CrewAI, or
custom state-machine engines.
Container Orchestration: Advanced
Kubernetes, Helm, Docker, and serverless
architectures for dynamic runtime scaling.
¢ Protocol Interoperability: Deep understanding
of Model Context Protocol (MCP) to link tools
and external APIs seamlessly.
¢ Inference Acceleration: Experience with model
hosting platforms (vLLM, TensorRT-LLM,
Ollama), batching strategies, and GPU resource
allocation.
¢ Memory Architecture: Designing persistent
short/long-term memory, semantic caching,
vector databases (Pinecone, pgvector), and
Knowledge Graphs.
¢ System Languages: Production-level code in
Python, Go, or Rust for performant orchestration
services.
Observability & AgentOps
Security & Execution Sandboxing
¢ Agent Evaluations (Evals): CI/CD regression
testing for non-deterministic multi-agent systems
using Ragas, DeepEval, or Braintrust.
¢ Runtime Sandboxing: Securing autonomous
code/tool execution via microVMs (Firecracker),
gVisor, or WASM environments.
¢ Tracing & Logging: Implementation of
LangSmith, Langfuse, Phoenix, or
OpenTelemetry for step-level latency, tool loops,
and costs.
¢ Guardrail Integration: Real-time safety policies
(NeMo Guardrails, Guardrails AI) for input/output
sanitization, loop prevention, and ne-grained
RBAC.
2. Key Job Responsibilities
Domain
Core Focus & Key Deliverable
Runtime Platform
Architect microservice platforms that execute, auto-scale, and maintain stateful
multi-agent workows reliably in production.
Developer Experience
Cost & Routing
Deliver internal SDKs, CLI utilities, starter templates, and API abstractions to
accelerate product team agent development.

Domain
Core Focus & Key Deliverable
Implement dynamic model routing (fallbacks, SLM vs. LLM orchestration) to
continuously optimize cost and latency.
Governance & Safety
Establish human-in-the-loop (HITL) execution controls, detailed audit logging, and
automated safety boundary enforcement.
3. Required Experience & Qualications
Experience Expectations
¢ Education: Bachelor's or Master's degree in Computer Science, Software Engineering, AI, or equivalent
practical experience.
¢ Mid-Level (3+ Years): Backend/Platform software background with at least 1+ years building production-grade
LLM or agent-driven systems.
¢ Senior / Lead (5+ to 8+ Years): Background in Distributed Systems, MLOps, or Cloud Engineering, with proven
experience scaling autonomous AI infrastructure.
¢ Preferred Backgrounds: Platform Engineering, MLOps/LLMOps, DevOps/SRE, or Microservices Architecture.
AI Agentic Platform Engineer Competency Prole ¢ Published Document