Senior Full-Stack Engineer (AI Products) (Remote - India)

What matters most

• Production-Grade AI Pipeline Development

  The role requires moving beyond simple LLM wrappers to building robust, agentic systems that utilize tool calling, evaluation, and monitoring. You must demonstrate the ability to design pipelines that are reliable, scalable, and capable of handling multi-step tasks in a live production environment.

• End-to-End Feature Ownership

  As part of a small, agile team, you are expected to own features from database schema design and backend logic to frontend implementation. This requires a high degree of autonomy and the ability to translate ambiguous customer requirements into functional, user-centric product features.

• Technical Communication with Stakeholders

  You will work directly with non-technical customers and stakeholders. A key priority is the ability to explain complex technical trade-offs, architectural decisions, and AI limitations in clear, accessible language to ensure alignment between product goals and engineering execution.

Likely challenges to think through

• Balancing Rapid Development with System Reliability

  In a startup environment, the pressure to ship quickly often conflicts with the need for stable, production-ready AI systems. Prepare by discussing your methodology for implementing automated testing, monitoring, and error handling in AI workflows to ensure quality does not suffer during rapid iteration.

• Managing Ambiguity in AI Product Requirements

  Because the platform involves complex publishing workflows, requirements may be open-ended. You should be prepared to describe how you proactively identify technical hurdles and define scope independently when clear specifications are not provided by stakeholders.

Topics to prepare

• Production AI Pipelines and Evaluation

  The role requires moving beyond simple LLM wrappers to robust, production-grade systems. Review strategies for implementing LLM evaluation frameworks (e.g., RAGAS or custom benchmarks) and monitoring systems to track latency, cost, and output quality in production environments.

• Agentic Systems and Tool Calling

  Focus on the conceptual design of agentic workflows where LLMs interact with external tools to perform multi-step tasks. Study patterns for structured output, error handling during tool execution, and maintaining state across complex agent interactions.

Skills to revise

• Full-Stack Python and FastAPI

  Practice building asynchronous APIs using FastAPI that integrate with PostgreSQL. Focus on designing clean, maintainable backend architectures that can handle the high-concurrency requirements of AI-powered publishing workflows.

• React and TypeScript Integration

  Develop a small-scale feature that requires complex state management in React using TypeScript. Ensure you can demonstrate how to handle real-time updates or streaming responses from AI backend services within the frontend.

• Containerization and Cloud Deployment

  Create a Dockerfile for a Python-based AI application and simulate a deployment process to a cloud environment (e.g., Azure). Focus on optimizing container size and managing environment variables for secure, production-ready deployments.

• Prepare a production-grade AI project walkthrough

  Select a past project that goes beyond a simple LLM wrapper. Prepare to explain your architecture for agentic workflows, specifically detailing how you implemented tool calling, managed state, and handled error recovery in a production environment.

• Document your AI pipeline evaluation and monitoring strategy

  Create a brief summary of how you have monitored LLM performance in production. Be ready to discuss the metrics you tracked, how you handled latency, and the specific evaluation frameworks or custom logic used to ensure output quality for end users.

• Review full-stack integration patterns with FastAPI and React

  Since the role requires end-to-end ownership, prepare to discuss how you structure data flow between a PostgreSQL database, a FastAPI backend, and a React/TypeScript frontend. Focus on how you maintain type safety and handle asynchronous AI tasks.

• Draft examples of technical communication with non-technical stakeholders

  Prepare two scenarios where you translated complex technical constraints into product decisions for non-technical users. Focus on how you explained trade-offs regarding AI model selection, cost, or latency to help stakeholders understand the product impact.

• Formulate questions about the publishing platform's AI architecture

  Develop thoughtful questions regarding how PT handles the specific challenges of AI-powered layout generation, such as how they manage context windows for large documents or how they integrate human-in-the-loop feedback into their automated pipelines.