Job Title

Backend Engineer– Build What's Next Location:Chandigarh/Mohali (On-site) Type:Full-time Level:Mid-level (2-4 years)We're Nomisma, and we're building AI products that actually matter and tools that solve real problems for real people. The next decade belongs to people who canbuildwith AI, not just talk about it. We're looking for engineers who see LLMs, vector databases, and fine-tuned models not as buzzwords, but as raw materials for creation.If you want to be the person who architects systems, who makes the tough calls, who owns outcomes, not just tickets then keep reading.First: Who You Are (The Human Part) We care more about how you think than what you've memorized. You are: →An owner, not a passenger.You see a problem through to the end. You don't wait for permission. You ship. →Brutally clear.Your code is clean. Your PRs tell a story. Your Slack messages don't need a decoder ring. →Pragmatic over perfect.You ship fast, measure what matters, and iterate. You know "good enough today" beats "perfect never." →Reliable as gravity.When you say you'll do something, it's done. You treat production like it matters—because it does. →Endlessly curious.You read papers. You try new tools. You share what you learn. You make the people around you better.We care about: Can you build systems that don't break? Can you make intelligent trade-offs? Will you level up this team?What You'll Build (The Tech Part) You'll work across the full backend stack—from API design to model fine-tuning to production observability. Here's what we expect you to know (or learn fast):Core Stack Backend: Python + one of FastAPI/Flask/Django Data: SQL (PostgreSQL/MySQL), NoSQL (MongoDB), Redis, S3 APIs: REST/GraphQL, clean schema design, authentication (JWT/OAuth2)AI/ML Tooling Frameworks: PyTorch or TensorFlow Hugging Face: Transformers, PEFT (LoRA), Datasets Embeddings & RAG: FAISS, Pinecone, Weaviate, Chroma LLM APIs: OpenAI, Anthropic, Cohere—or open-source alternativesWhat You'll Actually Do No fluff. Real work: Build production-grade APIs that frontend and mobile teams can rely on. Clean contracts. Fast responses. Solid error handling. Architect RAG pipelines from scratch—ingestion, chunking, embedding, retrieval, evaluation. Make themfast . Make themaccurate . Fine-tune models (LoRA, QLoRA, full fine-tuning) when pre-trained isn't enough. Optimize inference so it's cheap and responsive. Own observability. Logs, metrics, dashboards, alerts. You'll know when something breaks before your users do. Ship features end-to-end. Work with React/React Native devs to wire up AI features that feel magical to users. Write docs. Clear, kind documentation. Future you will thank present you. Review code. Give thoughtful feedback. Raise the bar.Must Haves ✓ 2-4 years building backend systems with Python ✓ Hands-on with FastAPI/Flask/Django ✓ Solid understanding of API design, SQL/ORM, and auth patterns ✓ Real experience with Hugging Face and at least one vector database ✓ Comfortable with Docker, CI/CD, and one major cloud provider ✓ Strong communication and documentation skillsHow to Apply: Apply via LinkedInoratnomisma.ai/career/. You can also email your resume and 2–3 repo/notebook links with subject:“Backend Engineer – Your Name.”