About Me
AI Engineer at Acolyte AI, specializing in multi-modal LLM systems and building production-grade AI solutions.
Hello!
I am Sharan Babu, an AI Engineer at Acolyte AI based in Texas. CS graduate from George Mason University, specializing in the Large Language Model and Generative AI space with 2+ years of hands-on experience building innovative LLM-powered solutions.
Currently, I work as an AI Engineer at Acolyte AI, where I architect and deploy custom RAG systems for multi-modal chatbots, develop prompt engineering strategies across text, voice, and avatar-based interfaces, and collaborate with engineering teams to create rich AI-driven product experiences.
My journey includes developing UserSubContext - an innovative context management system for LLMs, creating KirinEdit - specialized text editing techniques for LLMs, and building Pralok - an AI assistant featuring dynamic interface generation and advanced context management. I've also had the privilege of working at Amazon as a Business Intelligence Intern.
Skills & Tools
LLM & AI
- Large Language Models
- RAG Systems
- Context Management
- Prompt Engineering
- NLP
Development
- Python
- SQL
- API Design
- System Design
Data & Tools
- TigerGraph
- Semantic Search
- Vector Databases
- Git
- AWS
What I Do
LLM Development
Building sophisticated applications powered by Large Language Models, including context management systems, code generation workflows, and dynamic interface generation.
RAG Systems
Designing and implementing Retrieval Augmented Generation systems that combine semantic search with LLM capabilities for accurate, context-aware responses.
Information Retrieval
Building search engines and deduplication systems using graph databases, NLP models, and custom indexing pipelines for efficient data discovery.
AI Product Development
Creating meaningful AI-powered products and abstractions that solve real problems, from concept to production deployment.
Awards & Recognition
Let's Connect!
I would love to connect and discuss large language models, information retrieval techniques, and building meaningful products powered by LLMs.