SK.

Dublin, Ireland · Open to engineering opportunities

Sandeep
Konduru Raju.

Backend engineering. Applied AI.

I build reliable services and investigate how AI can make them more useful. Four years in enterprise software at Manhattan Associates, now pursuing an MSc in Artificial Intelligence at NCI Dublin.

View resume Explore projects
GitHub ↗LinkedIn ↗Email ↗

01 / Experience

Built for real operations.

Sep 2021 — Dec 2025

Bengaluru, India

Software Engineer

Manhattan Associates

Developed Java microservices and REST APIs for enterprise warehouse operations, with work spanning database performance, delivery automation and production support.

  • Improved query response time by 45% through schema and indexing changes across 50M+ records.
  • Shortened release cycles by 40% using Jenkins and Docker workflows.
  • Resolved 100+ production issues through debugging, root cause analysis and distributed tracing.

Rising Star · Q2 2022   /   Star of the Quarter · Q1 2023

02 / Selected work

Small systems. Clear decisions.

Personal projects and academic experiments, with implementation status made explicit.

Input validation API

Implemented prototype

SecureAgent

A testable input-validation service that evaluates raw prompts before sanitization. Three rule-based classifiers vote on safety, with overrides for high-risk patterns.

Includes structured JSON logs, a health endpoint and ten test cases. The published implementation uses rules rather than LLM classifiers.

Python · FastAPI · Pydantic · pytest

Explore source code

Incident service foundation

In progress

AI Incident Investigation Platform

Building an incident investigation platform, starting with a layered Spring Boot service for incident creation, severity, status and health checks.

Implemented: in-memory incident storage and service/controller tests. Planned: RAG investigation, persistent tenant isolation, Kafka processing and cloud deployment.

Java · Spring Boot · REST APIs · JUnit

Explore source code

MSc / AI evaluation

Prompt-injection detection

Compared a single-model baseline, a three-LLM voting ensemble and a sanitization pipeline on 100 balanced benign and adversarial prompts.

Measured precision, recall and F1, including the trade-off that sanitization can conceal attack signals. This academic experiment is separate from the rule-based SecureAgent API.

Python · Groq API · LLM evaluation · SQLite

03 / Toolkit

Engineering fundamentals,
with room to learn.

Backend
Java, Python, Spring Boot, FastAPI, REST APIs
Data
SQL, PostgreSQL, MySQL, MongoDB, Redis, Kafka
Delivery & reliability
Docker, Jenkins, AWS, Linux, Git, debugging, code reviews
AI projects
LLM APIs, prompt engineering, input validation, model evaluation

04 / Education

Learning with purpose.

MSc Artificial Intelligence

National College of Ireland, Dublin

January 2026 — December 2026 (expected)

B.E. Computer Science

Sir M Visvesvaraya Institute of Technology, Bengaluru

2017 — 2021 · CGPA 7.97/10

Let’s connect

Have a role or a problem
worth exploring?

Interested in backend, platform and applied AI engineering opportunities.