Applied AI FDE

About the Role

The Applied AI Forward Deployed Engineer (FDE) at PRISM is a full-time, on-site position based in Gurugram. This is a builder role that sits at the intersection of business and engineering. You will embed directly with business and operations teams to understand how work actually gets done, then design and ship AI agents that take that work off their plates. Your core mandate is to bring measurable efficiency to PRISM's business processes — Accounts Payable (AP), Accounts Receivable (AR), revenue management, reconciliation, escalation management, and other high-volume operational workflows.

Key Responsibilities

• Partner directly with finance, revenue, and operations stakeholders to break down business processes into problems that agents can solve, and prioritize by impact.
• Design, build, and deploy LLM-powered agents and workflows that handle real operational tasks end-to-end — e.g., matching and clearing invoices in AP/AR, flagging and resolving reconciliation breaks, routing and triaging escalations, and surfacing revenue-management actions.
• Integrate agents with existing systems of record (ERPs, billing, ticketing, data warehouses) so they can read, act, and write back safely into production workflows.
• Build the supporting scaffolding — tool/function calling, retrieval, evaluation harnesses, guardrails, and human-in-the-loop checkpoints — needed to make agents reliable enough for finance-grade work.
• Run experiments, define success metrics with business owners (time saved, error rates, cycle time, exceptions resolved), and monitor agent performance in the real world, iterating quickly.
• Own the full lifecycle of what you ship: performance, reliability, security, cost, documentation, and code quality.
• Act as a translator and force multiplier — turning ambiguous business pain points into deployed solutions, and codifying learnings so PRISM's agentic capabilities compound over time.

Requirements & Tech Stack

• Strong software engineering skills in languages such as Python, Java, or similar, with experience building production-grade services, APIs, and integrations.
• Hands-on experience building LLM-powered applications and, ideally, autonomous or semi-autonomous agents using frameworks such as LangChain, LangGraph, LlamaIndex, or equivalents.
• Solid grasp of generative AI and LLM fundamentals: prompt engineering, tool/function calling, retrieval-augmented generation (RAG), fine-tuning, and evaluation of model outputs.
• Experience designing agentic patterns — multi-step reasoning, tool use, orchestration, and human-in-the-loop review — and deploying them into real product or operational workflows.
• Comfort working with business and finance stakeholders: able to sit with a process owner, understand workflows like AP/AR, reconciliation, or escalations, and translate them into technical requirements.
• Knowledge of cloud platforms (AWS, GCP, or Azure), containerization (Docker, Kubernetes), and modern CI/CD practices.
• Strong communication skills.

Qualifications

• Bachelor's or Master's degree in Computer Science, Engineering, Data Science, or a related field, or equivalent practical experience.
• Experience with financial/operational systems (ERPs, billing, reconciliation tools) or in consumer internet, hospitality, or urban-tech products is a strong bonus.