Vinay Kumar Kummarikuntla profile photo 1 on Tal

Vinay Kumar Kummarikuntla

AI Engineer @ BuddyLoan | GenAI, LLMs, RAG & Multi-Agent Systems | LangGraph, LangChain, vLLM | Python, FastAPI, CI/CD, AWS, GCP

Bengaluru

4+ years of work experience

Buddy Loan
Oct 20241 yr 11 mosPresent

Bengaluru

coinearth technologies
Jan 20231 yr 10 mosOct 2024

Hyderabad

MountBlue Technologies
May 20228 mosDec 2022

Bengaluru, Karnataka, India

Vinay’s AI Stack

cursor
cursor
langfuse
langfuse
langgraph
langgraph
langsmith
langsmith
Qdrant
Qdrant
vLLM
vLLM

Education

Bosscoder Academy

Software Engineer

Bosscoder Academy

20242024
Kakatiya University, Warangal

Bachelor of Technology - BTech, Information Technology

Kakatiya University, Warangal

20193 yrs2022

Join Vinay and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Vinay Kumar Kummarikuntla on Tal

Headline

AI Engineer @ BuddyLoan | GenAI, LLMs, RAG & Multi-Agent Systems | LangGraph, LangChain, vLLM | Python, FastAPI, CI/CD, AWS, GCP

About

AI Engineer building production GenAI systems at BuddyLoan, where I design and ship LLM-powered platforms that combine reasoning, retrieval, and automation to solve real business problems. I've built a hybrid AI reasoning pipeline — LLMs, deterministic business logic, structured knowledge retrieval, conversation memory, and intelligent tool calling — that cut manual business-configuration review time from ~12 hours/week to ~1 hour/week. One core project: an intent-driven multi-agent platform (Communication, Knowledge, Operations, and Integration agents) that plans, validates, and executes lender policy and business-rule changes end-to-end — reading emails via Gmail, retrieving grounded policy answers via RAG, and applying validated updates to our rule engine (GoRules/BRE), orchestrated with LangGraph and an LLM planner + plan validator. I've also QLoRA fine-tuned and self-hosted an LLM serving stack (Qwen on vLLM/NVIDIA GPUs) processing roughly 800M tokens/month with continuous batching and prefix caching — keeping sensitive data on enterprise infrastructure. On the engineering side: FastAPI for serving, Pydantic for structured data contracts between agents, Langfuse for LLM observability (tracking token usage, latency, and errors in production), PySpark for large-scale data pipelines, and CI/CD for reliable, scalable deployment (~5K requests/min on production inference services). Before AI Engineering, I worked as a Software Engineer building backend systems (Python, REST APIs, PostgreSQL, MongoDB, Redis) — experience that shapes how I build AI systems today: production-grade, observable, and designed to actually ship. Open to AI Engineer / GenAI Engineer / LLM Engineer roles where I can keep building systems that go from prototype to production.

Location

Bengaluru

AI stack

cursor

langfuse

langgraph

langsmith

Qdrant

vLLM

Work experience

Artificial Intelligence Engineer at Buddy Loan, Oct 2024 - Present · 1 yr 11 mos, Bengaluru

Software Engineer at coinearth technologies, Jan 2023 - Oct 2024 · 1 yr 10 mos, Hyderabad

Software Development Engineer at MountBlue Technologies, May 2022 - Dec 2022 · 8 mos, Bengaluru, Karnataka, India

Education

Software Engineer, Bosscoder Academy

Bachelor of Technology - BTech, Information Technology, Kakatiya University, Warangal

Vinay Kumar Kummarikuntla - Artificial Intelligence Engineer in Bengaluru | Tal