Senior AI Engineer

Role Overview

Patrika Digital is hiring a builder-first, hands-on Senior AI Engineer and Architect to construct the intelligent infrastructure powering our next generation of news systems. You will own our technical AI roadmap end-to-end, evaluating build-vs-buy API integrations, designing code architectures, and establishing the safety framework for AI systems. Over time, you are expected to step up and build out a focused engineering unit.

Key Responsibilities & Core Problems to Solve

• AI Search Indexation & Attribution: Build the core structured data architecture, content feeds, and licensing-aware crawl configurations that secure Patrika’s placement and attribution within modern LLM search engines (ChatGPT, Gemini, Perplexity, Google AI Mode).
• AI-Assisted Newsroom Tooling: Design reliable editorial workflows featuring automated drafting, summarization, metadata tagging, and localization engines built to handle dialectal variations across the Hindi belt.
• Personalization Engine: Architect an advanced, internal next-article recommendations and application stream algorithm to expand session depth and user duration.
• System Risk Mitigation: Serve as a technical gatekeeper to reject requests that introduce trust or SEO risks, preventing mass AI-generated thin content or hallucinations.

Experience & Scale Requirements

• Must have shipped production-grade ML/AI infrastructure (not merely early prototypes) processing millions of daily API requests under strict cost and latency parameters.
• Demonstrated technical experience working with Hindi/Indic NLP systems or a proven capacity to master low-resource language training metrics.
• Clear experience navigating AI models inside publishing or content environments where quality signals heavily impact trust indices and SEO.

Critical Disqualifiers

• A purely academic or theoretical AI research background lacking production system shipping experience.
• Lack of awareness regarding the structural risks of generative models in high-authority publishing (e.g., treating maximized text generation volume as an unmitigated win).
• Lack of interest or baseline experience with vernacular-language linguistic realities (e.g., feeding Hindi copy through standard, un-optimized English pipelines).