Akshit Ireddy

Partnered with the research team to integrate audio-to-face transformer models for real-time lip-sync and facial animation, using Python LiveLink servers, ONNX C++ pipelines, and Unreal Engine integrations. Developed low-latency WebRTC and Pixel Streaming pipelines for real-time avatar communication and supported scalable backend/deployment workflows for real-time agentic AI systems.

Integrated LLM-powered agentic NPCs into Second Life and designed memory-aware multi-user group conversation systems for coherent, context-driven interactions. Replaced gRPC with WebRTC to reduce latency and built an AI NPC modding framework for Unreal Engine and other game engines.

Built interactive LLM-powered NPC systems using Cohere APIs with LangChain and vector memory architectures for persistent, context-aware dialogue.

Developed Data Odyssey, transforming Excel datasets into interactive comic narratives using GPT-3.5, LangChain, and multiple ML/visualization components. Built natural language interfaces that generated executable Python code for automated visualization and AI-driven storytelling.

Currently pursuing an online Master of Science in Data Science at IIIT Hyderabad (Apr 2026–2028).

Completed a B.Tech in Computer Science and Artificial Intelligence at Manipal Institute of Technology (Aug 2021–May 2025).