Platform Mesh Engineer

Job Description

Job Req ID: 58790

Posting Date: 26 May 2026

Location: Bengaluru

Salary: Competitive

About the role

The Platform Mesh Engineer is a specialist software engineering role responsible for building and operating the interconnected platform layer that orchestrates AI models, agents, tools, and services into a reliable, composable system.
This role is similar in level to a Software Engineering Specialist but differs in responsibilities and ways of working due to the AI‑native nature of delivery. Instead of focusing on product feature code, the Platform Mesh Engineer creates the agent orchestration and integration fabric—a “service mesh, but for AI”—that enables multiple agents to discover capabilities, hand off work safely, route requests to the right models, and operate with enterprise‑grade guardrails and observability.

What you’ll be doing

1) Agent Orchestration Layer (“Service Mesh for AI”)
• Design and implement the agent orchestration layer: how agents discover each other, call tools, delegate tasks, and hand off outputs across workflows.
• Define interaction patterns for multi‑agent collaboration, including sequencing, parallelisation, and controlled handoffs—supporting a “parallel mesh” delivery model.
2) MCP Servers, Tool Registries & Integration Adapters
• Maintain MCP (Model Context Protocol) servers, tool registries, and API adapters that give agents secure, governed access to enterprise integrations.
• Own lifecycle management for tool definitions: onboarding, versioning, deprecation, and compatibility across agent clients.
3) Model Routing Strategy (Task / Cost / Latency Optimisation)
• Own the model routing strategy for directing requests to the right model (local, cloud, fine‑tuned) based on task type, cost constraints, and latency requirements.
• Implement routing policies and abstractions that allow model providers to change without breaking agent workflows (model gateway approach).
4) Platform Guardrails & Resilience Engineering
• Implement platform‑wide guardrails such as rate limiting, fallback chains, circuit breakers, and safe‑fail mechanisms to prevent runaway agent loops and degraded user experiences.
• Build resilient patterns for dependency failure (timeouts, retries, bulkheads) and ensure graceful degradation.
5) Observability & Failure Mode Management
• Ensure deep observability across the AI platform:
o tracing agent reasoning chains and handoffs
o logging tool calls and outcomes
o surfacing failure modes (hallucination patterns, tool errors, routing misses, latency spikes)
• Provide dashboards/telemetry that engineering teams and leadership can use to monitor reliability, throughput, and risk hotspots.
6) Platform Standards, Governance & Cross‑Team Enablement
• Define and enforce platform standards for:
o agent/tool registration
o security boundaries
o protocol mediation and gateway rules
o integration onboarding patterns
• Partner with Pod Lead Engineers and product teams to embed the mesh into SDLC workflows and ensure consistent adoption.

What Makes This Role Different from a Traditional Platform Engineer
• You build a mesh for AI agents, not just microservices.
• Routing decisions are model‑aware, balancing quality, cost, and latency.
• Observability includes agent chains and tool calls, not only request traces.
• Guardrails must prevent unique AI risks like runaway loops and uncontrolled parallel execution.

Essential Skills / Experience

• Strong experience designing and operating distributed systems and platform services.
• Excellent API design and integration engineering skills (REST/gRPC/event driven patterns).
• Hands‑on experience with infrastructure‑as‑code and modern deployment pipelines.
• Strong experience with observability tooling (logging, metrics, tracing) and reliability patterns.

Desirable Skills / Experience

• Familiarity with agent frameworks (e.g., LangGraph, CrewAI or equivalent) and multi‑agent orchestration concepts.
• Understanding of MCP concepts and tool registry / gateway patterns.
• Experience with model gateways and model abstraction/routing layers.

Success Measures (What “Good” Looks Like)

• Agents can reliably discover and use tools/integrations through a governed registry layer.
• Platform demonstrates strong resilience: safe fallbacks, stable latency, controlled load.
• End‑to‑end traces exist for agent workflows and tool calls, enabling rapid diagnosis and continuous improvement.
• Model routing reduces cost and latency without reducing quality for critical workflows.

BT is the UK’s leading communications group and the holding company behind some of the country’s most recognised brands – including BT, EE, Openreach and Plusnet. Our purpose is as simple as it is ambitious: we connect for good. Our customers include consumers, small, medium and large businesses, public sector organisations and other communications providers.

BT’s role is about setting direction, unlocking value and creating the conditions for our brands and businesses to thrive.

Having come through the most capital-intensive phase of our fibre investment, our focus now is on what comes next – simplifying how we operate, using technology and AI to work smarter, and organising ourselves to serve customers better and grow sustainably. Group teams shape strategy, policy, brand, capital allocation and transformation, helping the whole organisation perform at its best.

We have a singular culture that unites all our people: we are customer-first challengers, who are committed, clear and connected. These behaviours unite us as one team to deliver for our colleagues, our customers, our stakeholders and the country. Joining BT means working at the heart of a business that matters to the UK, with the opportunity to shape decisions, influence outcomes and help set the future course of one of the country’s most important companies.