Aditya Khandelwal

I'm a backend engineer with 3+ years of experience, currently working as an SDE II on the applied AI team at Interview Kickstart. Right now, I am building agentic systems that resolve user support queries efficiently and give a delightful user experience.

Outside of work, I am exploring how memory can enable agents to become more intelligent over time—learning from interactions to provide increasingly personalized and effective user experiences.

Work

AI Chatbot with RAG

Interview Kickstart · 2024

Production chatbot with Pinecone and Botpress. Automated 35% of support queries, increased satisfaction 15%.

Semantic Search

Interview Kickstart · 2024

Vector search handling 10K+ weekly queries with AI agent for content enrichment. Eliminated 30 hours/week of manual work.

Performance Optimization

Interview Kickstart · 2024

Eliminated API bottleneck with Redis caching and Django signals. Reduced P99 latency 92% (5s→400ms).

Projects

Undercover Agent

Social Deduction Game - Play now

Multiplayer game where players identify the undercover agent through questions and voting. Built with real-time WebSocket communication.

Yadash

Memory Layer for AI Agents

Hybrid memory system with PostgreSQL and ChromaDB. Extracts memorable context with LLM pipeline, reduced prompt costs 60%.

Invest Shrine

Investment Platform

Django backend with GraphQL APIs for startup investments. Flutter mobile app for deal flow management.

Connect

Reach out to me on: akaditya05@gmail.com, GitHub · LinkedIn