Skip to content
open to roles · India & Gulf

Gogularamanaa S

Software Engineer·Agentic AI in production

I design and build AI agents that run in production: document understanding, retrieval and the backend services around them. I also lead the test automation that keeps them dependable.

now Chennai · Genworx
buildingAgentic document-intelligence platform: LangGraph agents, RAG on pgvector, Gemini & Claude
leadingOrg-wide test automation framework across Genworx teams
shippingProduction services on Google Cloud: FastAPI, gRPC, Cloud Run
Google Cloud CertifiedProfessional Cloud Developer
✓ valid to 2028
graph: whoami · run #0001 ● running
retry on fail input parse agents ×3 retrieve tests output
graph state
Tap a node to see what it does in my work
✻ tools I ship with
✻ 01 · about

Agents, and the engineering around them

I'm a software engineer at Genworx, where I build agentic AI systems that run in production: LLM agents and retrieval that turn contracts and invoices into structured data. I also led a team of two that built Genworx's org-wide test automation framework. An AI system is only as useful as it is dependable, so I work on both.

AI engineeringagents · retrieval · LLMs

  • LangGraph agents
  • RAG on pgvector
  • Gemini & Claude
  • Document AI
3agents in production

Automationtests · CI · standards

  • Playwright E2E
  • pytest
  • Vitest
  • Led a team of 3
15+modules automated
✻ 02 · work

Selected work

Each case study walks through the problem, the architecture, the decisions and trade-offs, what went wrong, and the result.

run #01 · genworx✻ AI

Document Intelligence Platform

A headless platform that reads contracts, invoices and almost any business document, and returns structured data.

pipeline
contract.pdf
invoice.pdf
scan.png
summarytablesimages
{
  "type": "invoice",
  "vendor": "Acme Ltd",
  "total": 1240.00,
  "lines": 6
}
10docs per batch, in parallel
3specialist agents
2LLM providers
run #02 · genworx✓ AUTOMATION

Org-wide Test Automation Framework

I led two engineers to build the framework every Genworx team now uses, with shared Playwright libraries and tagging standards.

coverage
modules under E2E test15/15 ✓
Vitestfrontend unit
pytestbackend unit
Playwrightend to end
shared libstag rules enforced2 projects
15+modules automated
2projects covered
3engineers, I led
run #03 · genworx◇ BACKEND

Gateway & Integration Layer

A backend-for-frontend that routes every request over gRPC to the right microservice, with authorization in one place.

routing
frontend
BFF🔒 Permit.io
authconnectorsdoc platforminvoice AP/AR
connectors → QuickBooks Online Zoho Books
4services routed over gRPC
2accounting integrations
1place for authz
run #04 · freelance◇ CLIENT

Lending Ledger App

An Android app for a lending business that records money lent to each customer and tracks interest owed.

app + architecture
Ledger ₹ due
Customer A₹ 1,200
Customer B₹ 850
Customer C₹ 2,400
Customer D₹ 600
DUMMY DATA
Android app · React Native
↓
FastAPI · Cloud Run
↓
PostgreSQL
100+customers managed
2weeks to ship
1engineer, end to end
private client case study →
also built
  • Invoice Automation for AP/AR · GenworxReal-time invoice processing on top of the document platform; 500 invoices in a single run for one client.
  • RFID door lock · College projectArduino Uno and Embedded C; card and tag authentication.
✻ 03 · stack

What I use, grouped by job

only tools I've used hands-on · hover for where

✻ ai engineering

8
LangGraph
LangChain
pgvector
Gemini
Claude
Vertex AI
Azure Doc Intel.
RAG

✓ automation & testing

6
Playwright
pytest
Vitest
E2E
Tagging
Shared libs

✓ production practices

6
Batch workers
Retries
AuthZ
gRPC contracts
E2E gating
Cloud Run

◇ backend

8
Python
FastAPI
gRPC
Permit.io
Microservices
Node.js
Express.js
REST

◇ cloud & data

6
Google Cloud
Cloud Run
PostgreSQL
MongoDB
MySQL
IAM

◇ frontend · mobile · integrations

5
Angular
React
React Native
QuickBooks Online
Zoho Books
✻ 04 · ask

Ask my agent

A small version of the systems I build at work: a tool-using RAG agent that answers only from this site, shows each step and cites its sources.

✻Shows its steps. Every tool call and source is visible.
[1]Cites sources. Click a citation to jump to that part of the site.
⇄Two depths. Recruiter mode is short; engineer mode goes into the architecture.
⚙Takes actions. It can hand you my resume or draft an email to me.
✓Evaluated. Answers are checked for grounding in CI, like the rest of the site.
portfolio agentonline · RAG over this site
Ask anything about my work.
Try a suggestion below.
✻ 05 · principles

How I build AI systems for production

A demo answers one question well. A production system has to answer the next ten thousand reliably. These are the habits I bring to that work, each one from something I've shipped.

01

Ground every answer

Agents answer from retrieved data, not from memory. Retrieval runs over pgvector, and outputs are structured so the next system can check them.

in practice → Document Intelligence Platform
02

Test before trust

Unit and end-to-end tests are part of the system, not an afterthought. Shared libraries and tag rules keep every team's tests consistent.

in practice → Test Automation Framework
03

Design for load and failure

Work runs in parallel batches with workers, so one slow document doesn't hold up the rest, and a failed step can be retried on its own.

in practice → 10 documents per batch
04

Clear boundaries

One gateway, typed gRPC contracts between services, and authorization in a single place, so each service can change without breaking the others.

in practice → Gateway & Integration Layer
05

Own it end to end

From the data model to deployment on Google Cloud, I'd rather understand the whole path than hand off a piece of it.

in practice → Lending Ledger App, shipped solo

“The interesting part of AI engineering starts after the demo works.”

— what I keep coming back to