BUILDING AT THE INTERSECTION OF AI & SOFTWARE AHMEDABAD, INDIA

Yash RathodAI & Software Engineer

Intelligence.Engineeredinto reality.

I build coordinated AI agents, document-aware applications, and the backend systems that connect them to real work.

Explore selected work View résumé

GitHub LinkedIn EXPLORE BELOW

ANATOMY OF AN AGENT SYSTEMFIG. 01

INTELLIGENCEwith structure.

Coordinator

Decomposes the task and delegates work to specialized workers.

CURRENT / 2025 — PRESENTe2m Solutions Associate AI Engineer PREVIOUS / 2024 — 2025MagnusMinds IT Solutions Software Developer Intern

01 / SELECTED ENGINEERING WORK

Built to solve.
Designed to be understood.

Explore the system, inspect the decisions, and follow the source. Professional work and personal projects, with their evidence clearly marked.

PROFESSIONAL WORK / e2m SOLUTIONS

Multi-agent orchestration.

One task. Specialized workers.
A human in control.

A coordinator that breaks down work, delegates to specialized agents, and keeps people in control of the execution.

Independent specialist workers Persistent context across handoffs Local & AWS deployment

PythonFastAPIMCPSuperMemoryReactAWS EC2

Read the engineering case studyImplementation described in the résumé · Client code is private

CONCEPTUAL ARCHITECTUREA task. A coordinated system.

Human task

Chief coordinatorDecompose & delegate

Frontend

Backend

DevOps

Shared contextMCP tools

Human oversight & intervention

The coordinator breaks down a human task and assigns it to specialized workers.

FROM THE WORKBENCH

Public source · RAG prototype

Document retrieval assistant

A document-aware assistant connecting local PDFs, websites, and Google Drive to a retrieval pipeline with conversation history.

A document-change hash reuses the FAISS index instead of rebuilding an unchanged corpus.

Python / FastAPI / LangChain / FAISS

Case study Source code

Public source · Agent dashboard

Multi Agent System

A real-time interface for agent health, sessions, memory, commands, and activity, backed by a WebSocket gateway connection.

Request correlation, timeouts, and reconnect backoff make gateway failures visible.

Next.js / TypeScript / Node.js / Express

Case study Source code

Personal project · Resume

Crypto futures trading system

A real-time trading system combining multi-timeframe signals, composite scoring, and ATR-based risk management.

Streaming data, multi-timeframe signal filtering, and ATR-based execution risk management.

Python / CoinDCX API / WebSocket / Pandas

Case study

02 / PROFESSIONAL EXPERIENCE

Where the work
meets the real world.

Backend engineering, client workflows, and AI systems with operational responsibility.

2025 — Present

e2m Solutions

Associate AI Engineer CURRENT ROLE

Turning AI workflows into working systems.

Agent orchestration, automation, and client-facing AI applications across local and cloud environments.

Coordinated specialist agents with shared memory and operator controls.
Built tenant-isolated social publishing with personalized content.
Connected audits and meeting transcripts to approved client workflows.

FastAPI / React / Supabase / n8n / AWS EC2Explore the orchestration work

2024 — 2025

MagnusMinds IT Solutions

Software Developer Intern

A foundation in data and backend engineering.

Natural-language database access, Azure SQL migration, and query optimization.

Exposed natural-language-to-SQL queries through FastAPI.
Migrated client databases with schema and integrity checks.
Rewrote slow SQL queries to address production bottlenecks.

Python / FastAPI / SQL Server / Azure SQL

03 / CONNECTED EXPERTISE

The model is one layer.
I build the system around it.

Capabilities backed by implementation. Follow each discipline into a project to see it in practice.

Intelligence.

Context, tools, coordination.

Multi-agent systems
RAG
LangChain / LangGraph
MCP
LLM tool use
NLP

Inside multi-agent orchestration

Infrastructure.

The systems behind the model.

Python / FastAPI
REST APIs
PostgreSQL / Supabase
Azure SQL
AWS / Azure
Docker / Git

Explore the retrieval backend

Interfaces.

Make complex systems usable.

React / TypeScript
Real-time dashboards
WebSocket
Human-in-the-loop
n8n workflows
Data visualization

See the monitoring interface

HOW I THINK ABOUT SYSTEMS

Keep context explicit. Make failures visible.
Give people a way to intervene.

04 / BEHIND THE BUILD

Curiosity is the starting point.
Engineering makes it useful.

Yash Rathod Ahmedabad, India

I work at the intersection of applied AI and software engineering. I care about what makes a system dependable: explicit context, observable failures, and an interface that people can understand.

My personal work also explores market data, signal classification, and automated trading—a way to think carefully about feedback loops and risk.

EDUCATION / 2026B.E. in Computer ScienceLJ Institute of Engineering and Technology

Visit the public workbench

CONTINUED LEARNINGCertifications & coursework

6 résumé-listed credentials

01Oracle Cloud Infrastructure Generative AI Certified Professional

02Machine Learning · DeepLearning.AI

03Machine Learning in Production

04AWS Cloud Technical Essentials

05Building Generative AI-Powered Applications with Python

06Database Structures and Management with MySQL · Meta

Listed in the supplied résumé. Verification links were not provided.

05 / START A CONVERSATION AHMEDABAD, INDIA

Let’s build
intelligent systems.

Have an engineering challenge, a role, or an idea?
I’d like to hear about it.

Email Yash

yashrathod1710@gmail.com

LinkedIn GitHub Download résumé