Available for Full-Time Roles & Collaborations

IIT KHARAGPUR GATE AIR 64

Hi, I'm Suman Kumar Raj

GenAI & Machine Learning Engineer

Dual Degree graduate from IIT Kharagpur building production-grade GenAI and machine learning systems. Experienced in architecting stateful multi-agent workflows with LangGraph and production RAG pipelines.

Explore Projects View Resume GitHub LinkedIn LeetCode

IIT KGP

About Me

My Journey & Passion

Coming from an engineering background at IIT Kharagpur, I enjoy tackling hard engineering challenges—whether it's orchestrating multi-agent systems with LangGraph, building deep learning classifiers for computer vision, or developing open-source Python packages to profile execution bottlenecks.

When I'm not writing code or building AI pipelines, you'll find me mentoring students in Data Structures & Algorithms, exploring new ideas in systems architecture, or pushing the boundaries of what large language models can automate.

Technical Arsenal

Core languages, frameworks, and production tools powering my AI & engineering projects.

Generative AI & LLMs

LangChain
LangGraph
Agentic AI
Memory & Multi-Agent Systems
LLM Orchestration, RAG, VectorDB
Tool & Function Calling
Structured Outputs, Evaluation

Machine & Deep Learning

Classification, Regression
Feature Engineering, EDA
Linear & Logistic Regression
Decision Trees, Random Forest
SVM, KNN, K-Means, PCA
Recommender Systems, NLP

Programming Languages

Python (Expert)
SQL (Advanced Queries & EDA)

Frameworks & Libraries

PyTorch
TensorFlow & Keras
Scikit-learn
NumPy & Pandas
Matplotlib & Seaborn
TensorBoard

Platforms & Tools

Docker
Git & GitHub
LangSmith
VS Code & Jupyter
Google Colab & Kaggle

Core Engineering & Math

Data Structures & Algorithms
DBMS & OOPs Principles
Probability, Statistics & Linear Algebra
MLOps Fundamentals

Featured Projects

Deep-dive implementations in Agentic AI, Computer Vision, Deep Learning, and Open-Source Software.

LangGraph • Multi-Agent

Multi-Agent Research Assistant

Built a multi-agent research system in LangGraph that decomposes complex topics into specialized analyst perspectives with human-in-the-loop refinement.
Implemented parallel information retrieval using Tavily and Wikipedia to synthesize context across diverse domains.
Designed a map-reduce workflow to parallelize analyst interviews and aggregate insights into structured research reports.

Python LangGraph LangChain OpenAI Tavily LangSmith

Stateful Memory • Routing

MyTaskManager

Engineered a stateful task-management AI agent in LangGraph maintaining short-term conversational context and structured long-term memory.
Architected distinct Profile, To-Do, and Instruction memories using Pydantic schemas for precise extraction and adaptive execution.
Implemented cyclic workflows and dynamic routing to manage memory updates before safely returning control to the primary agent.

Python LangGraph OpenAI Pydantic Trustcall

Computer Vision • PAD

Fingerprint Liveness Detection (PAD)

Created a Presentation Attack Detection (PAD) biometric verification pipeline using MobileNetV3-Small transfer learning.
Developed an end-to-end processing pipeline: 70/15/15 split, image augmentation, ImageNet normalization, and single-image inference.
Calibrated decision thresholds against a target 3% BPCER; evaluated with APCER, BPCER, ACER, EER, and ROC-AUC metrics.

Python MobileNetV3 TensorFlow OpenCV

Deep Learning • 0.93 F1-Score

LendingClub Risk Prediction

Built an end-to-end credit risk classification system on 396K+ loan records with missing value handling, categorical encoding, and MinMax scaling.
Trained a deep neural network with ReLU activations, dropout regularization, and sigmoid output to classify loan default risk.
Attained a 0.93 F1-score on the test set while effectively resolving severe class imbalance.

Python TensorFlow Scikit-learn Pandas

Open-Source Python Package • PyPI

CodeBeat

Created and published an open-source Python profiling package leveraging Object-Oriented Programming (OOPs) principles for modular line-by-line function tracing and runtime statistics.

$ pip install codebeat

PyPI GitHub

Python OOPs Performance Profiling PyPI Distribution

All GitHub Repositories

Live API

Dynamically synchronized with @SKR18156592.

View on GitHub

Experience & Mentorship

Professional roles, academic positions, and teaching experience.

AI/ML Engineer Intern • MakeMyBrain

May 2024 – Jul 2024

Built an end-to-end mood-based music recommendation RAG system, processing structured metadata, acoustic features, and textual profiles across audio catalogs.
Engineered a hybrid search pipeline combining BM25 keyword matching and dense vector embeddings (α = 0.4), boosting retrieval Recall@10 from 0.61 to 0.79 compared to lexical search alone.
Implemented a cross-encoder re-ranking stage to filter top 100 retrieved candidates down to top 10–20 tracks before LLM generation, ensuring personalized and context-aware recommendations.

Internship

DSA Mentor • AlgoZenith

Oct 2022 – May 2024

Mentored 500+ students in Data Structures and Algorithms, directly improving their problem-solving capabilities and interview success rates.
Guided participants through rigorous coding challenges, time-complexity optimization techniques, and structured interview-focused problem-solving frameworks.

500+ Students

Teaching Assistant • Indian Institute of Technology, Kharagpur

Aug 2023 – May 2025

Assisted in conducting tutorials and evaluating assignments for undergraduate courses at IIT Kharagpur.
Facilitated doubt-clearing sessions, helping students build stronger understanding of core concepts and analytical techniques.

IIT Kharagpur

Education

IIT Kharagpur

CGPA: 8.12

Dual Degree (B.Tech + M.Tech) in Ocean Eng. & Naval Architecture

2020 – 2025

S K P Vidya Vihar, Bhagalpur

92%

Class XII (CBSE)

2020

D P S Sabour, Bhagalpur

10 CGPA

Class X (CBSE)

2018

Awards & Achievements

GATE 2024 — AIR 64

Ranked in the 99.8th percentile nationwide in the Graduate Aptitude Test in Engineering.

Prof. J.P. Ghose Memorial Award

Received the prestigious Prof. J.P. Ghose Memorial Award (2022–2023) at IIT Kharagpur for exceptional academic performance.

Gold Medalist — Gymkhana Championship

Gold Medal winner in Choreography at IIT Kharagpur (2021–2022).

Verified Credentials

Licenses & Certifications

Specialized training, advanced coursework, and technical certifications.

DeepLearning.AI Issued August 19, 2026

Retrieval-Augmented Generation (RAG)

Authorized through Coursera

Completed advanced training on building production-grade RAG pipelines, focusing on chunking methodologies, dense/sparse embeddings, and hybrid search mechanics to optimize LLM groundedness and factual accuracy.

Hybrid Search Chunking Strategies Embeddings Vector Databases

Verify Certificate

LangChain Academy Issued August 4, 2026

Foundation: Introduction to LangGraph - Python

Authorized through LangChain Academy

Completed foundational training on LangGraph, mastering state management, graph compilation, routing, and tool integration to architect resilient, stateful multi-agent AI systems and production-grade workflows in Python.

LangGraph State Management Multi-Agent Systems Python

Verify Certificate

Udemy Issued May 20, 2026

Python with Data Structures & Algorithms

Instructors: Krish Naik, Mayank Aggarwal, KRISHAI Technologies

Completed an intensive bootcamp focusing on advanced Python programming and core Data Structures & Algorithms, backed by rigorous LeetCode coding exercises.

Python Data Structures & Algorithms LeetCode

Verify Certificate

Udemy Issued May 27, 2021

Python for Data Science & Machine Learning

Instructors: Jose Portilla, Pierian Training

Completed foundational training covering data analysis libraries, numerical computing, data visualization, and core machine learning algorithms using Python.

Data Science Machine Learning Python

Verify Certificate

Extra-Curricular Activities

Community engagement, social service, and campus involvement.

NSS Volunteer • Unit 9, IIT Kharagpur

2+ Years of Community Service

Served as a dedicated volunteer for over 2 years in NSS, Unit 9, IIT Kharagpur.
Contributed to social welfare initiatives, community development campaigns, and campus outreach programs.

IIT Kharagpur

Get in Touch

Feel free to reach out for collaborations on agentic systems, deep learning pipelines, or software roles.

Email sumanraj4176@gmail.com

Phone +91-9798196180

GitHub SKR18156592

LinkedIn sumanraj11

LeetCode sumanraj112002