Forward Deployed AI Engineer
JLL
Bengaluru, Karnataka, India3h agovia LinkedIn
About the Role
JLL is deploying AI agents directly into the business to automate workflows, support decision-making, and integrate with existing systems. As a one-person engineering team for your assigned business area, you will leverage the methodology, platform, and patterns provided by the central AI team's Center of Excellence. Your primary function is to translate business problems and SMEs' input into shipped agents and MCP integrations, handing off mature work to platform and operations teams. Engineers at this level are expected to integrate AI tools as a primary component of their workflow, using provided hooks, skills, workflows, and automations for safe production deployment. Leverage in this role is derived from AI, not headcount.
Key Responsibilities
- Build and ship agents for the business in close partnership with business SMEs and operators.
- Stand up MCP servers and connectors, moving from pre-production to production through CDR.
- Bootstrap skills, integrations, and evals on top of JLL's AI platform (including LLM gateway, shared RAG services, and MCP toolchain).
- Run workshops with business teams, teach them to use Claude Code, and translate their workflows into agent skills.
- Hand mature systems off to platform and operations teams once they outgrow the bootstrap phase.
- Feed patterns, anti-patterns, and platform gaps back into the Center of Excellence.
Requirements & Tech Stack
- Shipped production agents using Claude Agent SDK, MCP, or an equivalent stack.
- Production code in at least one of Python, Node, or .NET, with comfort moving across these languages.
- Experience working with diverse data stores, including:
- Relational
- NoSQL
- Object/blob
- Vector
- Document
- Operated end-to-end as a one-person team, covering design, shipping, support, and hand-off.
- AI tools are expected to be a primary part of your workflow.
Qualifications
- Strong written and verbal communication skills; comfortable in discussions with non-engineers.
- Pragmatic judgment regarding when to ship, harden, or hand off work.
- Systems thinking with a focus on failure modes.
- Comfort with tasking and the ability to run multiple agents in parallel as part of your daily workflow.