About this role
Enterprises struggle to reliably orchestrate business processes in production using AI at scale because existing systems lack a unified, governed execution layer between LLM providers and enterprise infrastructure. Workato addresses this challenge by building an enterprise AI platform that unifies data, applications, and processes while developing foundational infrastructure like the MCP Gateway and AI Gateway. As a Senior AI Engineer based in Singapore, this person will design and build high-throughput, low-latency network services, implement protocol-level features such as request routing and tool-call proxying, and optimize PostgreSQL data layers for concurrency and performance. This role exists to strengthen the connective tissue between enterprise systems and AI agents, requiring a pragmatic engineer who balances deep systems-level knowledge with a rigorous approach to production ownership and code quality.
What you will do
- Design and develop the MCP Gateway and AI Gateway as production services that mediate between applications, AI agents, and LLM providers.
- Build high-throughput and low-latency network services handling TCP, TLS, HTTP/1.1 and HTTP/2, JSON streaming, and connection pooling.
- Own the application-side data layer by writing and optimizing complex PostgreSQL queries, schemas, and indexing strategies.
- Manage concurrency through worker pools, queues, graceful shutdowns, and race-free shared state while profiling services under load.
- Drive comprehensive observability for AI systems by tracking token usage, latency per provider, cache hit rates, and operational costs.
What you need
- Bring senior-level experience of at least 5 years building and operating production services in Go, Ruby, or both.
- Possess deep application-developer knowledge of PostgreSQL, including query optimization, indexing, transactions, and migrations at scale.
- Demonstrate practiced skill in concurrency, performance profiling, and debugging production bottlenecks using tools like pprof and flamegraphs.
- Show familiarity with Kubernetes and containerized environments, covering resource limits, health checks, and rolling deployments.
- Understand LLMs at the protocol level, including message structures, tool calling, streaming responses, and direct API integrations.
Nice to have
- Familiarity with Model Context Protocol (MCP) including building or integrating MCP servers and clients
Skills and tech stack
Questions candidates ask
What kind of systems will I be building in this role?
You will build the infrastructure layer that connects enterprise systems to AI, specifically focusing on an MCP Gateway, an AI Gateway, and surrounding services. These systems sit between LLM providers and enterprise applications to handle routing, authentication, rate limiting, observability, and protocol translation.
What technical background is required for Go or Ruby?
Candidates need senior-level experience shipping production services in Go, Ruby, or both. For Go, this means deep familiarity with the standard library and performance optimization, while Ruby candidates should have strong production Rails experience and know how to tune ActiveRecord.
How does the team approach the use of AI tools in development?
The company encourages but never forces the use of AI and LLM tools in development to help engineers work faster. Team members have access to major tools and models with generous usage limits, though developers remain fully accountable for everything they merge and must be able to justify all architectural choices.