|
I build production-grade AI systems, agent architectures, and backend pipelines. Currently studying Artificial Intelligence & Data Science at Sri Shakthi Institute of Engineering and Technology, Coimbatore.
Sri Shakthi Institute of Engineering and Technology, Coimbatore, India
I'm Shreekumar B — an Artificial Intelligence & Data Science student at Sri Shakthi Institute of Engineering and Technology. I build production-grade AI systems that solve real problems — not demos, not toy projects.
From audit intelligence platforms for CA firms to autonomous company operation systems, I architect end-to-end — from database schema to AI pipeline to UI.
Focused on developing production-grade backends, pipelines, and autonomous workflows.
An AI-native operating layer for companies. Automates repetitive workflows using agent routing, structured LLM decision nodes, and state-machine transactional persistence.
Full-stack ingestion engine for scanning and analyzing large codebases. Builds structured semantic layout indexes and provides natural language queries using Gemini API.
Hackathon MVP. Reconciles sales registers against e-way bills and government tax portals to detect GST mismatches, missing invoices, and critical compliance gaps.
Computer vision & multi-modal LLM architecture designed for automating examination paper evaluations. Scans physical handwriting, matches answer patterns against structured rubrics, and generates scored mark lists.
Developer intelligence platform integrating execution chains for automated directory modification.
Real-time object detection and alerting pipeline connected directly to live RTMP camera feeds.
A modular interface system supplying layout components and tokens for React implementations.
Academic research milestones and projects completed during 2026.
Designed agent orchestrations and transactional state evaluations for enterprise workflows.
Managed development of continuous AI audit checker pipelines matching invoicing tables.
Architected codebase scanning engines utilizing parallel token processing and indexing trees.
Programmed post-OCR validation chains and rubric score checkers matching physical papers.
Released initial directory semantic agent parsing file structures directly on command execution.
Pursuing specialized studies in Artificial Intelligence & Data Science, focusing on databases and modular system designs.
Available for systems design, AI backend integrations, or automated agent pipelines.