Available for Hire

AI Backend Engineer

I build production AI systems and scalable backend infrastructure. My core work is Python, Django, and FastAPI — with deep experience in RAG pipelines, LLM integration, and async processing. I've shipped AI assistants with zero-cost inference architecture, optimized APIs by 40-60%, and built systems handling real traffic on AWS and Azure.

Let's Talk View All Projects

Core Expertise

Python

Django

FastAPI

RAG Pipelines

LLM Integration (OpenAI, Gemini, Llama)

Qdrant

PostgreSQL

Celery/Redis

AWS

Elasticsearch

Relevant Projects

A selection of my work that demonstrates my capabilities as a AI Backend Engineer.

AI-Powered Portfolio with RAG Assistant

Portfolio site with a production RAG chatbot (Qdrant + sentence-transformers + multi-provider LLM with fallback). CI/CD pipeline auto-generates LaTeX resume PDFs from shared JSON data via GitHub Actions.

Astro React TypeScript

HealthLab Management Platform

Pathology lab management system with Go/Gin backend and React admin dashboard. Appointment scheduling with capacity tracking, digital report delivery, and analytics.

Go (Gin) PostgreSQL React

Message Scheduler

Schedule messages via email and Telegram. Django backend with Celery/Redis task processing, cron-based execution, and natural language date parsing.

Python Django Celery

Frequently Asked Questions

What AI systems has Ankit built in production?

Ankit has built a production RAG assistant at ankitjang.one using Qdrant vector search, sentence-transformers embeddings, and multi-provider LLM fallback (Gemini/OpenAI/Llama) with zero-cost inference. At Saara Inc, he led SalesGPT — an AI sales assistant for e-commerce built on OpenAI GPT-3.5/GPT-4 with 40% cost optimization through caching and request throttling.

Can Ankit optimize existing backend systems?

Yes. Ankit specializes in backend performance optimization — led a company-wide Django-to-Jinja migration reducing page load from ~1000ms to <200ms (5x improvement), optimized APIs by 40-60% via N+1 elimination, index additions, and Celery/Redis async offloading. He writes about these techniques in his blog on backend performance optimization.

Ready to work together?

I'm currently available for remote opportunities. Let's discuss how I can help bring your vision to life.

Email Me Connect on LinkedIn