
Nilay Mallik
agentic ai builder
AI that agents: I design and build e2e AI systems to reduce cost and improve quality
yap more
I design and develop multi agent workflows, RAG pipelines, LLM powered applications, and production style AI products that are reliable, scalable, and impactful.
My work spans from intelligent agent design and tool augmented reasoning to vector search, knowledge retrieval, and e2e AI product development.
I’m passionate about turning ideas into production ready systems that create meaningful impact. Currently seeking agentic AI builder / AI engineer / any similar opportunities where I can learn, build, and contribute.
Tech Stack
the tech arsenal behind my builds!
projects
All repositoriesNilora
Agentic AI chatbot
A production-style conversational agent with tool routing, multi-format RAG, persistent memory, SSE streaming, and speech input.
activity
@nilaymallikkDeepSeek Harness: The Future of AI Agents?
Most AI coding agents look similar from the outside. You give them a prompt. They read your repository. They edit files, run commands, search for thin...
Read articleMemory Architecture AI Agents Need
Agentic AI Memory Architecture The first version of an AI assistant often has a surprisingly simple memory architecture. Take every previous message,...
Read articleDesigning RAG for 100 Million Documents
A RAG demo is easy. Take a PDF, split it into chunks, generate embeddings, put them in a vector database, retrieve the five nearest chunks, attach the...
Read articleLoop Engineering Is Dead? Why Graph Engineering Is the Next Evolution of AI Systems
The phrase “Loop Engineering is dead” is memorable, but it is technically wrong. Loop Engineering is not disappearing. It is becoming a building block...
Read article






