AI Full-Stack Engineer

AI Full-Stack Engineer (Backend Heavy)

Location: Mumbai | Work from Office

Experience: 3-5 Years

Employment: Full-time

Preferred Background: IITs / BITS Pilani or equivalent strong engineering background

IMPORTANT

Please complete the screening form for us to understand about you in detail: https://forms.gle/Faf2S13wTSGGwxJw7

About Discovr AI

Discovr AI is building an AI-native infrastructure layer for creator advertising, enabling brands and agencies to automate creator discovery, campaign execution, content intelligence, brand safety, measurement, and reporting.

We are building production-grade AI systems and agentic workflows that operate across large volumes of creator, campaign, content, audience, and brand data.

We’re looking for an AI Full-Stack Engineer who can own systems end-to-end, with deep strength in backend engineering, system architecture, and AI infrastructure.

This is a highly hands-on, startup engineering role.

What You’ll Build

• Build and scale backend systems using Golang and/or Python.
• Design APIs, services, data pipelines, and distributed backend architectures.
• Build production-grade AI infrastructure and LLM-powered applications.
• Develop agentic workflows, RAG pipelines, retrieval systems, tool integrations, and AI automation.
• Integrate LLMs with databases, APIs, internal tools, and business workflows.
• Build scalable data systems using MongoDB and related databases.
• Develop product-facing interfaces using TypeScript and modern frontend frameworks.
• Own features end-to-end from architecture and backend to frontend and production deployment.
• Improve system performance, reliability, latency, scalability, and AI inference costs.
• Build evaluation, monitoring, observability, and guardrail systems for AI applications.
• Solve complex engineering and architecture problems as the product scales rapidly.

What We’re Looking For

• 3–5 years of strong software/product engineering experience.
• Strong backend engineering expertise in Golang and/or Python.
• Hands-on experience building AI/LLM infrastructure in a B2B SaaS or AI product company.
• Strong experience with MongoDB, databases, APIs, and backend systems.
• Working proficiency with TypeScript and modern frontend development.
• Experience with LLM APIs, AI agents, RAG, embeddings, vector databases, or orchestration frameworks.
• Strong understanding of system design, data structures, algorithms, and software architecture.
• Ability to write clean, scalable, production-quality code.
• Experience taking products or AI prototypes from 0→1 and into production.
• Strong debugging and problem-solving ability.
• Comfortable working with high ownership and ambiguity in a fast-moving startup.

Strongly Preferred

• Engineering background from IITs / BITS Pilani.
• Experience at an AI-first B2B SaaS startup.
• Strong GitHub, open-source, competitive programming, or personal project track record.
• Experience with LangGraph, LangChain or similar agent frameworks.
• Experience with OpenAI, Gemini, Claude or other foundation models.
• Experience with vector databases, retrieval infrastructure and AI evaluation.
• Exposure to cloud infrastructure, Docker, Kubernetes, queues, caching, and distributed systems.
• Strong DSA / competitive programming fundamentals.

The Kind of Engineer We Want

You are not a frontend engineer who occasionally touches APIs, and you are not someone who has only experimented with LLM wrappers.

You should be comfortable going from:

Product problem → architecture → backend → AI layer → database → frontend → deployment → iteration

We’re looking for builders who enjoy solving hard engineering problems, shipping quickly, and taking complete ownership of what they build.

Why Discovr AI?

• Build AI infrastructure and agentic systems from the ground up.
• Work on real production AI problems, not proof-of-concept demos.
• High ownership across product, architecture, and engineering.
• Work closely with Product, Engineering, and Founders.
• Solve complex problems across AI, advertising, creators, and large-scale data.
• Build in a fast-moving 0→1 / 1→10 startup environment.