Founding Full Stack Engineer · Ascend Cargo

Gopi Sri Krishna Yarlagadda

I build the AI-powered freight and logistics SaaS platform at Ascend Cargo from scratch: scalable Python and Java microservices on Kubernetes, secure authentication, and agentic AI with RAG and the Model Context Protocol that automates shipment and invoice workflows.

See my work Get in touch
Portrait of Gopi Sri Krishna Yarlagadda
Dallas, TexasAgentic AI · RAG · MCP
90%less manual labor from automated FTP and file workflows
35%better login efficiency and security controls (OAuth 2.0, JWT, RBAC, SSO)
30%faster API response times through distributed caching
10tools in my solarnetwork-mcp server for solar asset management
About

Turning ambiguous problems into production software.

I lead full stack work end to end, from the data pipeline and the API to the agent that sits on top of it.

Agentic AI, RAG and MCP

I build agents that do real work: shipment and invoice agents that read freight data, retrieve the right context and automate multi-step workflows. I also wrote and open-sourced an MCP server so any AI assistant can query solar telemetry and diagnose faults.

Platform engineering

Microservices and APIs in Python and Java, deployed on Kubernetes with autoscaling, built to stay up under high-volume ingest.

Auth and security

OAuth 2.0, JWT, RBAC, Microsoft Entra ID and Google SSO.

Reliability

Designated Responsible Individual for backend systems: monitoring, on-call and faster restoration.

Delivery

CI/CD with GitHub Actions and observability tooling for faster issue detection.

Selected work

Things I have built and shipped.

Depth over volume: two systems I can defend line by line, then a few earlier projects.

Public repository · MCP server

solarnetwork-mcp

An MCP server that lets AI agents read solar telemetry, find dated equipment faults, estimate lost energy and write a technician work order. Its rule: it says “not assessed” when the data can’t support an answer, instead of guessing “healthy”.

  • 10 tools from fleet discovery to PDF service reports
  • Peer-relative fault detection across a fleet of public SolarNetwork nodes
  • Honest by design: handles gaps, rate limits and missing data explicitly
TypeScriptMCPstdioPDF reports
Read the write-up GitHub
AI assistantClaude · any MCP client solarnetwork-mcp10 tools · stdio SolarNetworkpublic API Diagnosisfaults · lost kWh Work orderPDF for the tech
Full stack · Agentic AI · RAG · MCP

Ascend Cargo AI logistics platform

Founding full stack engineer on a freight and logistics SaaS built from the ground up. One platform spans rates and carriers, orders and shipments, gate and yard operations, billing and audit, analytics and AI agents. My work includes agentic features using RAG and MCP integrations to automate shipment and invoice workflows.

Company site
Web application

Weather Application

A location-based weather app using the Dark Sky and Mapbox APIs to surface temperature, humidity and chance of rain for trip planning.

GitHub
Web application · JSP

College Administrative System

Certificate management that generates Provisional Certificates and Consolidated Marks Memos directly from student records.

GitHub
Experience

Where I have done the work.

Founding Full Stack Engineer

Ascend Cargo
  • Building an AI-powered freight and logistics SaaS platform from scratch, spanning rates and carriers, orders and shipments, gate and yard operations, billing and audit, analytics and AI agents.
  • Designed and operated scalable microservices and backend APIs for high-volume data ingest in Python and Java, deployed on Kubernetes with autoscaling.
  • Built agentic AI features, including RAG and MCP integrations, powering shipment and invoice agents that automate freight workflows.
  • Implemented secure authentication (OAuth 2.0, JWT, RBAC, Microsoft Entra ID, Google SSO), improving login efficiency and security controls by 35%.
  • Served as DRI for backend systems, monitoring availability and leading on-call efforts to cut restoration times.
  • Automated end-to-end FTP file workflows triggered via REST APIs and service bus patterns, cutting manual labor by 90%.
  • Optimized distributed REST APIs with caching, improving response times by 30%; built CI/CD (GitHub Actions) and cloud-native observability tooling.

Software Developer

Virtusa
  • Engineered features and refactored the “Grievance Cell” web product with Java and Spring Boot, reducing issue resolution time by 40%.
  • Delivered maintainable Java modules for ticket assignment and user feedback (Spring Boot, REST APIs, JUnit), improving delivery speed and reducing defect leakage.
  • Worked across a cross-functional team on uptime and rapid restoration, raising user satisfaction by 35%.

Research Developer

Dhanekula Institute of Engineering and Technology
  • Developed distributed JSP modules for college administration, automating student certification workflows.
  • Added monitoring, reporting and automated testing for certificate issuance, reducing errors and manual handling by 90%.
  • Found and fixed security vulnerabilities, including XSS, in Java Server Pages modules.
Skills

The toolbox, grouped by what it does.

Agentic AI and GenAI

  • Agentic AI development
  • Retrieval-Augmented Generation (RAG)
  • Model Context Protocol (MCP)
  • AI workflow automation

Languages

  • Python
  • Java
  • JavaScript
  • TypeScript
  • SQL
  • C++

Frameworks

  • React
  • Spring Boot
  • Hibernate
  • JSP
  • PHP

Cloud and DevOps

  • Google Cloud
  • Docker
  • Kubernetes
  • Linux
  • GitHub Actions
  • Git

APIs, data and testing

  • REST APIs
  • MySQL
  • Flyway
  • Postman
  • JUnit
  • Selenium

Auth and security

  • OAuth 2.0
  • JWT
  • Microsoft Entra ID
  • Google SSO
  • RBAC
  • XSS mitigation

Practices

  • Agile (Scrum)
  • SDLC
  • On-call and DRI
  • Jira
  • Confluence
Writing

Notes from the build.

The same article lives on this site and on Medium.

Featured article

An AI agent that diagnoses solar faults, and says “not assessed”

How I built an MCP server that lets AI agents read solar telemetry, find equipment faults and refuse to say “healthy” when they cannot know. Architecture, the ten tools and real findings from a live node.

Read on this site Read on Medium

Machine learning notes

Titanic: Machine Learning from DisasterMedium ↗ Caltech101: Airplanes, Motorbikes and SchoonersMedium ↗ Text classification on the Ford Sentence datasetMedium ↗ Image classification on wild animalsMedium ↗
Education

Where I learned it.

2022 – 2024

M.S. Computer Science

The University of Texas at Arlington · GPA 3.55
Data Structures, Algorithms, Software Engineering, Cloud Computing, Web Development

2018 – 2022

B.Tech Computer Science and Engineering

Dhanekula Institute of Engineering and Technology · GPA 3.3

2016 – 2018

Board of Intermediate Education

Sri Chaitanya Educational Institutions, Andhra Pradesh · Aggregate 86.6%

  • MicrosoftAzure Fundamentals AZ-900
  • CiscoNetworking Essentials · View badge
  • CourseraObject Oriented Programming in Java
  • NPTELThe Joy of Computing using Python
  • NPTELProgramming in Java
Contact

Let’s build something useful.

I’m always glad to talk about agentic AI, platform engineering or interesting problems in logistics software.

gopisrikrishna.y@gmail.com Download resume
LinkedIn GitHub Medium

Away from the keyboard: gardening, cooking, music and exploring new places.