AI-First Backend Develope

AI-First Backend Developer
Company: Vakta AI
Experience: 2+ Years
Employment Type: Full-Time
Location: Remote / Hybrid – India
Compensation: Industry Standard
About the Role
Vakta AI is building a real-time conversational AI tutor powered by voice AI, LLMs, RAG, document intelligence, and personalized learning.
Our MVP is already ready, and we are now looking for an AI-First Backend Developer who can help take the product from MVP → Production → Scale.
You will work on the core backend and AI infrastructure, improve the existing system, make it production-ready, and help scale it as user traffic grows.
Key Responsibilities
Build and maintain backend services using Python / FastAPI.
Improve and productionize the existing MVP architecture.
Build REST APIs, WebSockets, async services, and background jobs.
Work on real-time AI pipelines involving STT, LLM, RAG, TTS, and streaming.
Integrate LLMs, embeddings, vector databases, rerankers, and AI APIs.
Optimize MongoDB/PostgreSQL, Redis, caching, and database queries.
Improve system latency, reliability, security, and performance.
Set up monitoring, logging, error tracking, testing, and CI/CD.
Deploy and manage services on cloud infrastructure.
Identify bottlenecks and scale backend services as usage grows.
Optimize AI infrastructure for performance, scalability, and cost.
Debug production issues and take ownership of backend systems.
Required Skills
2+ years of backend development experience.
Strong knowledge of Python.
Hands-on experience with FastAPI, Django, Flask, or similar frameworks.
Experience with REST APIs, WebSockets, and async programming.
Experience with MongoDB/PostgreSQL/MySQL.
Experience with Redis.
Familiarity with Docker and cloud platforms such as AWS/GCP/Azure.
Strong debugging and problem-solving skills.
Understanding of backend architecture, security, and scalability.
AI-First Skills
Experience with some of the following is preferred:
LLM APIs
RAG
Embeddings
Vector databases
Semantic / hybrid search
Reranking
STT / TTS
Voice AI
Streaming LLM responses
Open-source AI models
AI agents / tool calling
Experience with tools such as Milvus, Qdrant, LiveKit, WebRTC, Hugging Face, Ollama, vLLM, LangChain, Prometheus, Grafana, or Sentry is a plus.
What You’ll Own
A major part of this role is:
Understand the existing MVP → Make it production-ready → Launch → Monitor → Optimize → Scale
We are looking for someone who can work with an existing codebase, improve what is already built, solve real production issues, and help scale Vakta AI as the number of users and concurrent AI sessions increase.
Why Join Vakta AI?
Work on a real AI product already moving toward production.
Build real-time conversational and voice AI infrastructure.
Work with LLMs, RAG, STT/TTS, and scalable backend systems.
High ownership and direct impact on product architecture.
Work closely with the founding and engineering team.
Fast learning and growth in an AI-first startup.
Apply with: Resume + GitHub/Portfolio + relevant backend/AI projects.

Work arrangement
No