Hi, my name is

Suchith Koduru

I build intelligent AI systems.

AI Software Engineer specializing in LLM agents, multi-agent architectures, and memory-augmented AI systems. Currently building AI agents at 10xscale.ai.

01. About Me

Hello! I'm Suchith, an AI Software Engineer passionate about building intelligent systems that push the boundaries of what's possible with Large Language Models and autonomous agents.

I recently graduated from the Indian Institute of Technology Delhi with a degree in Computer Science Engineering. My journey into AI started with curiosity about how machines can reason, remember, and collaborate—leading me to specialize in:

  • Building multi-agent systems with LangChain & LangGraph
  • Implementing persistent memory in AI agents (Qdrant, Mem0)
  • Creating MCP servers for database-aware AI agents
  • Contributing to open-source AI frameworks

I'm currently focused on LLM fine-tuning, exploring ways to make AI agents more capable, contextually aware, and reliable for real-world applications.

Profile Photo

02. Experience

AI Software Engineer

@ 10xscale.ai

June 2025 — Present
  • Contributed to the open-source agent framework Agentflow, with PRs #17 and #19 successfully merged
  • Built a Model Context Protocol (MCP) server for PostgreSQL, enabling natural language querying of structured data with database-aware agents
  • Designed multi-agent architectures using LangChain, LangGraph, and FastMCP with contextual memory and scalable execution
  • Implemented persistent memory systems using Qdrant vector database and Mem0 for improved long-horizon reasoning
  • Created a robotic prototype powered by Raspberry Pi with real-time AI-generated responses for live demonstrations
Python LangChain LangGraph FastMCP Qdrant Docker PostgreSQL

Research Intern

@ IIT Delhi (Prof. Kapil Tomar)

May 2024 — July 2024
  • Developed a full-stack Maternal Health Risk Predictor integrating ML with a responsive web interface for real-time inference
  • Benchmarked multiple models (SVM, RF, XGBoost, Gradient Boosting), achieving 99% accuracy with Gradient Boosting
  • Designed and deployed a Flask-based application with prediction visualization and interpretable output highlighting
Python Flask Scikit-learn XGBoost HTML/CSS/JS

03. Projects

Other Noteworthy Projects

MCP PostgreSQL Server

Model Context Protocol server enabling natural language querying of PostgreSQL databases. Integrated with Agentflow for database-aware agent capabilities.

  • Python
  • FastMCP
  • PostgreSQL
  • SQL

Customer Churn Prediction

End-to-end ML pipeline for telecom churn prediction using Random Forest with SMOTE, SHAP interpretability, and threshold optimization for business objectives.

  • Python
  • Scikit-learn
  • SHAP
  • Pandas

CIFAR-100 Classification

High-confidence image classification using WideResNet (WRN-28-10) with Sharpness-Aware Minimization (SAM) and temperature scaling for calibrated predictions.

  • PyTorch
  • WideResNet
  • SAM
  • NumPy

Drum Kit

Interactive web-based drum kit that responds to keyboard inputs and on-screen button clicks. Fun project showcasing DOM manipulation and audio APIs.

  • JavaScript
  • HTML
  • CSS
  • Web Audio

Simon Game

Classic memory game with progressively challenging sequences. Tests pattern recognition and memory skills with colorful visual feedback.

  • JavaScript
  • jQuery
  • HTML
  • CSS

Academic Database System

PostgreSQL-based academic administration database inspired by IIT Delhi's internal system. Features stored functions, triggers, and auto-update mechanisms.

  • PostgreSQL
  • SQL
  • Database Design

04. Skills & Technologies

Languages

Python C++ C JavaScript SQL

AI/ML Frameworks

LangChain LangGraph LiteLLM FastMCP PyTorch Scikit-learn

Databases & Vector Stores

PostgreSQL MySQL Qdrant Mem0

Tools & Infrastructure

Git Docker Linux GitHub Copilot LaTeX

Web Development

HTML CSS JavaScript Flask

05. Education & Achievements

B.Tech in Computer Science Engineering

Indian Institute of Technology Delhi

2021 — 2025

CPI: 6.36/10.0

Scholastic Achievements

AIR 479
JEE Advanced 2021

Among 1.2 lakh shortlisted candidates

AIR 620
JEE Mains 2021

Among 10 lakh+ candidates

Rank 201
TS-EAMCET 2021

Among 2.2 lakh students

06. Get In Touch

I'm currently open to new opportunities and collaborations. Whether you have a question, want to discuss AI projects, or just want to say hi, feel free to reach out!

Say Hello