12 Sep
|
Pupiltree.ai
|
Karnataka
12 Sep
Pupiltree.ai
Karnataka
AI-FIRST WEB GAME ENGINEER - GAME GENERATION & QUALITY AUTOMATION Pupiltree.ai | Full-time | On-site - Ballari, Karnataka ABOUT THE ROLE Pupiltree.ai is building a platform that can create web-based games in bulk from reusable templates. We are looking for a hands-on engineer who can build engaging browser-game templates, integrate them into our AI-powered generation harness, and create the automated quality-control systems that make generated games reliable, testable, and continuously better. This is not a one-game-at-a-time development role. You will help build the factory: reusable templates, standard interfaces, generation workflows, validation rules, automated tests, release gates, feedback loops, and the golden dataset that allow game production to scale without sacrificing quality. WORKING ARRANGEMENT - PLEASE READ This is a permanent, full-time position and requires compulsory on-site work in Ballari (Bellary), Karnataka. It is not a remote or hybrid role. Food and accommodation are provided. Specific arrangements will be explained during the hiring process. WHAT YOU WILL DO • Review our existing web games, templates, generation workflow, and cloud harness. • Design and build reusable templates for multiple types of browser-based games. • Create new games and variants with clear rules, responsive controls, reliable state management, and polished player feedback. • Define standard interfaces for configuration, assets, levels, scoring, difficulty, win/loss states, telemetry, and lifecycle events. • Integrate templates into the generation harness using Claude Code and/or the Claude Agent SDK, MCP integrations, cloud SDKs, APIs, storage, queues, deployment workflows, and CI/CD where appropriate. • Make generation runs versioned and reproducible across prompts, configurations, templates, models, and code revisions. • Build automated checks for compilation, runtime errors, gameplay rules, scoring, state transitions, responsiveness, performance, assets, and browser compatibility.
• Develop programmatic playtests, simulations, test agents, or deterministic replay systems where they improve confidence. • Create an automated QC and evaluation layer that scores each generated game and blocks outputs that fail agreed thresholds. • Capture structured reviewer feedback, player signals, defects, and accepted outputs. • Convert validated feedback into regression tests, improved evaluation criteria, stronger templates, better generation instructions, and updates to the golden dataset. • Add observability and reporting so the team can understand failure patterns and quality trends across batches. • Document the template specification, integration workflow, QC framework, and operating procedures. WHAT SUCCESS LOOKS LIKE • New game templates can be added through a clear, repeatable process. • Generated games are automatically built, launched, exercised, evaluated, and accepted or rejected. • Failures are traceable and easy to diagnose. • Human and player feedback becomes structured data that improves future templates, tests, evaluations, prompts, and the golden dataset. • Game output can scale without manual review effort increasing at the same rate. REQUIRED EXPERIENCE • Strong experience building web-based games or highly interactive browser applications. • Proficiency in TypeScript/JavaScript and contemporary web-development fundamentals. • Experience with Phaser, PixiJS, Three.js, Canvas/WebGL, or an equivalent browser-game or rendering stack. • Ability to design reusable architectures, template systems, and clean integration contracts.
• Experience with automated browser testing and programmatic validation using Playwright, Cypress, or similar tools. • Experience building cloud-based workflows using SDKs, APIs, storage, queues, containers, serverless services, or CI/CD systems. • Strong understanding of game state, input handling, scoring, progression, difficulty, deterministic behaviour, and edge cases. • Ability to translate quality goals into measurable checks and useful evaluation criteria. • Strong debugging, documentation, and end-to-end ownership skills. NICE TO HAVE • Production experience with Claude Code, the Claude Agent SDK, MCP, or equivalent coding-agent workflows. • Experience with LLM-assisted code generation, AI agents, evaluation harnesses, or synthetic-data workflows. • Experience designing golden datasets, regression corpora, rubric-based evaluations, or human-in-the-loop review systems. • Knowledge of telemetry, experimentation, player analytics, or automated gameplay simulation. • Familiarity with procedural content generation and configuration-driven game systems. • Experience optimizing browser games for performance, accessibility, touch devices, and multiple screen sizes. • Experience operating large batch pipelines with traceability, retries, idempotency, and cost controls. HOW TO APPLY Please apply with your resume or LinkedIn profile and include: • Links to browser games, interactive products, developer tools, or generation systems you have built. • A short description of the most relevant game architecture or automation pipeline you owned. • An example of how you tested game logic or evaluated generated software or content at scale. Code samples, playable demos, technical write-ups, and architecture diagrams are welcome. Please explain how you would use AI-powered agents and automated QC to generate and validate many games without manually reviewing every output.
📌 Game Developer (Karnataka)
🏢 Pupiltree.ai
📍 Karnataka