Senior Full-Stack Product Engineer — Agentic AI Applications

Agivant Technologies; Hyderabad; On-site; INR 1,300,000 - 1,800,000; Full time

Job summary

Agivant Technologies is building a small, high-velocity engineering team focused on productized agentic AI applications on Savanna. The Senior Full-Stack Product Engineer will own the user experience, architecture, security, and deployment, turning AI capabilities into a complete enterprise product.

The role involves end-to-end development across the stack, ensuring scalability, observability, and configurable architectures for multiple customers and prospects.

Qualifications

• Strong professional software engineering experience.
• Strong experience building modern, intuitive customer-facing UI/UX and translating complex capabilities into simple user workflows.
• Demonstrated ability to work across the application stack and take ownership from initial concept through deployment and production support.

Responsibilities

• Design and build intuitive, production-quality user experiences for agentic AI applications running on Savanna.
• Translate complex agent, graph, and AI capabilities into workflows that are understandable and useful to business users.
• Build interactive investigation experiences, visualizations, and human-in-the-loop workflows that allow users to understand what an agent is doing, review its recommendations, and approve, reject, modify, or override actions.
• Work directly with product, customers, prospects, sales engineering, and solutions teams to rapidly prototype, test, and improve user workflows.
• Create reusable UI/UX patterns and application components that can be applied across Fraud, AML, KYC, cybersecurity, and future agentic solutions.

Skills

Software engineering

UI/UX design

Cloud deployments

Productization

Enterprise applications

Tools

AWS

GCP

LLM orchestration

Job description

Role Overview

We are building a small, high-velocity engineering team focused on developing productized agentic AI applications on top of Savanna. Our initial application will focus on agentic fraud investigation, with the broader goal of creating a reusable library of industry and functional applications that can be rapidly configured, deployed, and commercialized across customers and prospects.

Role Overview

We are building a small, high-velocity engineering team focused on developing productized agentic AI applications on top of Savanna. Our initial application will focus on agentic fraud investigation, with the broader goal of creating a reusable library of industry and functional applications that can be rapidly configured, deployed, and commercialized across customers and prospects.

We are looking for a Senior Full-Stack Product Engineer who will work alongside an AI Engineer focused primarily on agents, LLM orchestration, GraphRAG, tool use, and AI/backend capabilities.

This role will own much of what turns those AI capabilities into a complete enterprise product: the user experience, application architecture, security and permissions, agent safeguards, deployment, scalability, observability, and commercial packaging.

This is not simply a frontend role. The successful candidate should be able to take an application from concept through production, work across the full application stack when needed, and make pragmatic decisions with limited direction.

Responsibilities

Product & User Experience

• Design and build intuitive, production-quality user experiences for agentic AI applications running on Savanna.
• Translate complex agent, graph, and AI capabilities into workflows that are understandable and useful to business users such as fraud investigators, analysts, and operations teams.
• Build interactive investigation experiences, visualizations, and human-in-the-loop workflows that allow users to understand what an agent is doing, review its recommendations, and approve, reject, modify, or override actions.
• Work directly with product, customers, prospects, sales engineering, and solutions teams to rapidly prototype, test, and improve user workflows.
• Create reusable UI/UX patterns and application components that can be applied across Fraud, AML, KYC, cybersecurity, and future agentic solutions.

Full-Stack Application Development & Productization

• Build and maintain modern customer-facing applications, including the user interface, application services, integrations, and supporting backend capabilities.
• Design applications that support long-running and asynchronous agentic workflows, real-time status and results, user and workflow state, and integration with external enterprise systems.
• Create modular, configurable architectures that separate reusable product capabilities from customer-specific configuration and integrations.
• Build applications that can move efficiently from prototype to demo, pilot, and production without being rebuilt for each customer.
• Maintain a strong standard for software quality, testing, reliability, maintainability, and production readiness.

Agent Safety, Security & Enterprise Controls

• Design safeguards that prevent agents from taking unauthorized, dangerous, unintended, or excessively costly actions.
• Implement appropriate human approval checkpoints, permissions, execution boundaries, and controls for sensitive or high-impact actions.
• Ensure agents act only within the authority of the user and application, with clear separation between read-only activities and actions that modify data or external systems.
• Provide mechanisms to monitor, pause, cancel, or terminate agent executions and prevent runaway or uncontrolled behavior.
• Build complete auditability and traceability for agent activity, including user requests, decisions, tool use, approvals, actions, results, and errors.
• Work with the AI Engineer to incorporate evaluation, validation, and policy checks before consequential actions are executed.
• Build applications suitable for security-conscious enterprise customers, including regulated industries.

Deployment, Scale & Cloud Marketplaces

• Own the production deployment and operational readiness of agentic applications across AWS and Google Cloud, as well as other customer deployment environments where required.
• Design applications to scale reliably across users, concurrent agent workflows, workloads, and customer deployments, with the monitoring and operational controls needed to support them in production.
• Productize and package agentic applications for commercial distribution through AWS Marketplace and Google Cloud Marketplace.
• Create repeatable deployment and upgrade approaches so marketplace-delivered applications can be provisioned, configured, monitored, supported, and maintained with minimum manual effort.

Requirements

What We’re Looking For

• Strong professional software engineering experience building and shipping production applications.
• Strong experience building modern, intuitive customer-facing UI/UX and translating complex capabilities into simple user workflows.
• Demonstrated ability to work across the application stack and take ownership from initial concept through deployment and production support.
• Experience building scalable, secure, reliable enterprise applications and operating them in public cloud environments.
• Experience turning prototypes or customer-specific solutions into configurable, reusable products.
• A working understanding of agentic AI applications and the challenges created when AI systems can take actions, not just generate answers.
• Strong product judgment and empathy for both technical and non-technical users.
• Comfort working through ambiguity, making pragmatic decisions, and moving quickly with limited direction.
• Strong collaboration and communication skills, including the ability to work directly with customers, prospects, product, engineering, and go-to-market teams.

Preferred Experience

• Experience with agentic AI applications, graph-based applications or visualizations, investigation or case-management workflows, cloud marketplaces, multi-tenant SaaS, regulated enterprise environments, or fraud/AML/KYC/cybersecurity is valuable but not required.

What Success Looks Like

• Turn AI and graph capabilities into polished applications that customers can understand, trust, and use.
• Own major parts of the product experience, application architecture, safeguards, deployment, and production readiness.
• Make agent behavior transparent, controllable, authorized, and auditable.
• Build reusable products that can be deployed repeatedly across customers rather than rebuilt as one-off solutions.
• Help move solutions from prototype to production and ultimately to commercial distribution through AWS and Google Cloud marketplaces.