Margati Inc

This is a 6-month paid internship designed as a path to full-time employment. You'll work on complex, high-visibility systems from day one, solving problems where AI and engineering meet real-world dataTop performers will be offered a full-time role with compensation and equity. Responsibilities Build intelligent scraping pipelines with fallback logic (GET Playwright LLM). Parse messy HTML using DOM-aware logic to extract emails, names, and keywords. Use lexical similarity and subsequence scoring to associate entities with goals. Integrate and optimize LLM APIs (OpenAI, Claude, Sonar) for reasoning and enrichment. Create structured reports in JSON/HTML from semi-structured sources. Optimize backend performance for multi-threaded, concurrent web crawling. Experiment with goal-classification models, AI-driven contact curation, and relevance ranking. Requirements A strong academic background in CS, engineering, AI, or related fields. Solid programming skills in Python (and optionally Node.js ). Direct experience with LLMs, including usage of OpenAI/Claude APIs in real projects. Projects or work demonstrating hands-on use of AI for reasoning, enrichment, or extraction. Deep curiosity and creativity in solving open-ended, data-heavy problems. The ability to move fast, think clearly, and work hard. A willingness to iterate, debug, and own real production features. Bonus If You Also Have Experience scraping the web at scale using Playwright, asyncio, or similar. Understanding of DOM traversal and proximity-based entity matching. Built tools or algorithms using subsequences and string similarity measures. Strong opinions on when and how to use AI to supplement human reasoning. This job was posted by Kishore Vaidyanathan from Margati.

This is a 6-month paid internship designed as a path to full-time employment. You will work on complex, high-visibility systems from day one, solving problems where AI and engineering meet real-world data. Top performers will be offered a full-time role with compensation and equity. Responsibilities - Build intelligent scraping pipelines with fallback logic (GET Playwright LLM). - Parse messy HTML using DOM-aware logic to extract emails, names, and keywords. - Use lexical similarity and subsequence scoring to associate entities with goals. - Integrate and optimize LLM APIs (OpenAI, Claude, Sonar) for reasoning and enrichment. - Create structured reports in JSON/HTML from semi-structured sources. - Optimize backend performance for multi-threaded, concurrent web crawling. - Experiment with goal-classification models, AI-driven contact curation, and relevance ranking. Requirements - A strong academic background in CS, engineering, AI, or related fields. - Solid programming skills in Python (and optionally Node.js). - Direct experience with LLMs, including usage of OpenAI/Claude APIs in real projects. - Projects or work demonstrating hands-on use of AI for reasoning, enrichment, or extraction. - Deep curiosity and creativity in solving open-ended, data-heavy problems. - The ability to move fast, think clearly, and work hard. - A willingness to iterate, debug, and own real production features. Bonus If You Also Have - Experience scraping the web at scale using Playwright, asyncio, or similar. - Understanding of DOM traversal and proximity-based entity matching. - Built tools or algorithms using subsequences and string similarity measures. - Strong opinions on when and how to use AI to supplement human reasoning.,