Open to opportunities

Kurapati Abhiram

@kurapatiabhiram

I build production agentic LLM systems, Indic NLP features, and low-latency AI inference platforms.

At Coinearth Technologies, I've shipped 14 production AI features for a WhatsApp-style social messaging platform across chat, voice, video, NLP, speech, and computer vision. I build agentic systems with LangChain and LangGraph, including ToDoZee, a voice assistant that executes six user actions end-to-end through tool calling.

I've built live translation for 19 Indian languages using IndicTrans2, fine-tuned Qwen, Gemma, and T5 models, and delivered RAG, call summarization, OCR, document-reading, and object-detection pipelines. I also co-authored a Springer-published paper on deep-learning methods for early Alzheimer's detection.

Experience

Work history, roles, and key accomplishments

Current

AI Engineer

Current

Coinearth Technologies

Dec 2025 - Present (8 months)

Shipped 14 production AI features for a social messaging platform, including generative AI, NLP, speech, and computer vision. Built task-oriented AI agents with tool calling, fine-tuned models, and implemented RAG and evaluation pipelines.

Systems Engineer

Infosys

Aug 2025 - Dec 2025 (4 months)

Completed Phase 1 of the Infosys foundation training programme, covering core Java and MySQL with hands-on assessments. Exited during Phase 1, prior to stream allocation.

Java MySQL

Full Stack Developer Intern

NUNC Systems

Aug 2023 - Oct 2023 (2 months)

Delivered 6+ web features in React for the LawLearning platform.

React

Education

Degrees, certifications, and relevant coursework

Vellore Institute of Technology, Chennai

Bachelor of Technology, Computer Science (AI & ML)

2021 - 2025

Grade: 7.20/10

Pursued a Bachelor of Technology in Computer Science with a specialization in Artificial Intelligence and Machine Learning, achieving a CGPA of 7.20/10.