40%

Complete your profile to find better job opportunities

Full Stack AI Engineer at Supernova AI, HSR

Full Stack AI Engineer

Supernova AI, HSR, Bengaluru

About the Company

Supernova builds AI that helps the world communicate: India's largest consumer AI startup, with 15M ARR, 10M+ Play Store downloads (4.7 rated, Top 10 on Play Store and App Store) and Superflow at 1M+ downloads. Backed by KAE Capital, Lumikai and founders from Swiggy, 1MG and Uni. Mission: help people communicate clearly and confidently across languages, give everyone an AI communication partner, and unite coaching and real-time assistance on one shared AI platform – English first, built for a global audience.

About the Role

Build AI-powered communication experiences using foundation models and voice AI: speech-to-clear-writing, cross-language communication and feedback on expression. Combine product engineering with prompting, evaluation and reliable AI systems. Prototype fast, learn from real usage, and improve quality, latency and cost on the shared platform. Much is still undefined, so you'll experiment and make calls without set answers.

Key Responsibilities

  • Design and prototype AI features, build demos/PoCs, iterate on feedback
  • Craft prompts; version prompts and model configs to trace regressions; manage context (chunking, metadata, retrieval)
  • Build hybrid rule-based + AI systems, robust pipelines for non-deterministic output, guardrails, fallbacks and human-in-the-loop workflows
  • Optimise performance and cost for API or hosted models
  • Build eval sets from real conversations, edge cases and production failures; define criteria for correctness, clarity, tone, usefulness, latency and cost; combine automated checks with human review; compare prompts and models over repeated runs
  • Monitor production for drift/regressions; turn failures into eval cases
  • Assess new models, explore fine-tuning/hosting, share learnings

Technical Requirements

Engineering: work across the full stack; quickly model unfamiliar systems (architecture, data flow, failure modes); explain and defend your design choices; review agent-written code as rigorously as your own; avoid needless abstraction and fragile design; open small, focused PRs.

AI: built products where LLMs drive the UX; know model tradeoffs; design prompts/context for consistent output; evaluate systematically and diagnose failures; monitor AI features in production.

Backend: HTTP, DNS, proxies; relational DB design; APIs, queues, cron jobs. Plus: Docker, Kubernetes, AWS/GCP/DigitalOcean, WebSockets.

Frontend: strong JS, HTML, CSS fundamentals; JavaScript/TypeScript with React, Vue or Svelte; responsive design.

Strong debugging and root-cause analysis.

How We Work with AI

AI agents write most of our code; engineers direct it and own the result. You must be able to explain the architecture, data flow, failure modes and trade-offs of anything you ship, including behaviour under unusual or malicious input. Test and simplify before a PR – never pass unchecked agent output to reviewers. Prototypes move fast; production must be excellent: edge cases, copy, loading/failure states and performance. Shipping starts the feedback loop. Every team uses AI (designers/PMs prototype; Devin in Slack lets anyone explore and fix code); you'll help improve these shared tools.

Qualities We Love

  • Strong engineering judgment
  • Moves fast and finishes well
  • Cares about craft
  • Relentless about improvement
  • Delivers checked, ready-to-use work
  • Ambitious, eager to grow
  • Contributes beyond assigned tasks
  • Empathetic toward users

Why Join

  • Cutting-edge work
  • Talented team (avg. age ~25)
  • Competitive salary, generous ESOPs, relocation support
  • Bangalore office with direct founder mentorship

Values: Move Fast, Break Things • Radical Candor • Super Experimentative Culture • Strong Opinions, Loosely Held