Headline

AI Engineer, Production LLM Systems | RAG, LangGraph, Agentic Workflows | 1B+ Tokens/Month | 1,500+ Hours Saved/Year | Ex-RingCentral

About

I build production AI systems that turn complex domain problems into reliable software end-to-end, from database to Infra and deployed LLM pipeline. Currently building agentic systems at Future Standard. I think and write about production LLM systems, RAG architecture, and the gap between AI demos and shipped software. Always happy to chat AI engineering — DM open. At OnFinance AI, I built ComplianceOS end-to-end (DB + backend + cloud + LLM pipelines) for BFSI compliance automation. Designed a production RAG pipeline over financial regulations and engineered LangGraph + LiteLLM agentic workflows handling 1B+ tokens/month. Contributed to AI features that helped drive $800K+ ARR and supported the $4.2M Pre-Series A with technical demos. At RingCentral, I modernized legacy Angular 1.x into React + Redux for enterprise contact-center products — cutting load time from ~2s to <0.5s and raising test coverage to 95%. Same end-to-end ownership, different domain.

Location

Bengaluru

AI stack

deepseek

hermes agent

langchain

langgraph

openai

PydanticAI

Work experience

Analyst - AI & Agent Systems at Future Standard, Jun 2026 - Present · 3 mos, Bengaluru, Karnataka, India, • Designing agentic memory architectures that let AI agents retain and recall operational context across multi-step workflows. • Building agent systems for operations and process intelligence — automating discovery, decision support, and execution monitoring for operational workflows. • Owning end-to-end delivery: agent orchestration (LangGraph/LiteLLM), backend APIs, evaluation pipelines, and production deployment.

AI and Full Stack Engineer at OnFinance AI, Jun 2025 - May 2026 · 1 yr, Bengaluru, Karnataka, India, • Built ComplianceOS end-to-end (DB → backend → cloud → LLM pipelines), cutting BFSI compliance effort by 1,500+ hours/year • Shipped AI features that contributed to $800K+ ARR growth • Designed a production RAG pipeline for complex financial regulations with versioned embeddings and citation grounding • Engineered LangGraph + LiteLLM agentic workflows processing 1B+ tokens/month for BFSI compliance products • Built technical demos used in $4.2M Pre-Series A raise.

SDE Intern at RingCentral, Jan 2025 - Jul 2025 · 7 mos, Bengaluru, Karnataka, India, • Migrated critical components from Angular 1.x → React + Redux, modernizing legacy systems • Improved application performance by 3x (2s → <0.5s load time) • Increased test coverage to 95% using React Testing Library • Implemented accessibility improvements for enterprise contact center systems • Integrated features into CI/CD pipelines ensuring production reliability

Education

Bachelor of Engineering - BE, Computer Science, RV College of Engineering

Pre University, Science, Mahesh PU college Mangalore

Public links

github: https://github.com/KTS-o7

leetcode: https://leetcode.com/KTS-o7