Rishabh Singh
Data Scientist | AI Engineer | IIT Madras
I'm all about that magic!
Data Scientist with hands-on experience delivering end-to-end analytics solutions — from data wrangling and modeling to building dashboards, APIs, and deployable prototypes. Worked on forecasting, mixed-effects sizing, MMM, and time-series pipelines with stakeholder-ready insight delivery.
Strong in Python, SQL, Streamlit, Flask, Figma, and cloud exposure (Azure/AWS), with experience mentoring junior analysts and collaborating across Forecasting, Product, and I&A teams.
Featuring an interactive CI-vs-PI regression simulator and merged contributions to CPython, conda-build & librosa.
IIT Madras BS student? I've put together notes, resources & guidance from my own journey.
Explore the guide →
About Me
I'm Rishabh Singh
Data Scientist & AI EngineerRishabh Singh is a Data Scientist and AI Engineer at Chryselys, building end-to-end analytics solutions — forecasting pipelines, mixed-effects models, Marketing Mix Modeling, and time-series systems with stakeholder-ready delivery. He graduated from IIT Madras with a B.S. in Data Science and Applications (2021–2025) and a Minor in Economics & Finance.
His open-source contributions include merged pull requests to CPython (Python's core interpreter), conda-build, and librosa — demonstrating systems-level Python expertise alongside applied ML work. Technical stack: Python, pandas, scikit-learn, Streamlit, Flask, FastAPI, SQL, Azure, AWS, Databricks, Dataiku, Redis, and MLOps. He has built and deployed full-stack ML applications including Meridian, an AI meeting intelligence platform, a live parking management system, and an LLM YouTube metadata pipeline, and writes about AI and machine learning on Medium with 23 published technical articles.
- Degree:Data Science and Applications
- Location:Delhi, India
- Leadership:oneArena Flask / Machine Learning Trainer
- Competition:Rank-3 Codechef Competition
Skills
Languages
ML & Analytics
Frameworks & Tools
Cloud & Infra
Education
B.S. Data Science and Applications
Indian Institute of Technology, Madras (IITM)
CGPA: 7.89
Minor in Economics & Finance
Indian Institute of Technology, Madras (IITM)
Experience
ML Intern @ iNeuron.ai | Cybersecurity, ML Model
- •scikit-learn, restAPI, Azure
- •Predicts whether domains are real or malicious
- •Enhancing cyber-security.
- •Machine Learning.
Data Scientist @ Chryselys | Data Science & ML
- •Machine Learning and AWS
- •Insight-Driven Analysis
- •Product Development
- •APLD Data Expertise
Services
What I'm Doing?
Back-end Development
Specialized in architecting, developing, and deploying robust and scalable back-end systems using Flask and Django frameworks.
Machine Learning Apps
Proficient in creating interactive Machine Learning Applications and visualizations. Using scikit-learn, PyTorch and other famous ML Tools.
Research in Machine Learning
Passionate about exploring advanced topics and staying updated with the latest methodologies in Machine Learning research.
Self-learning
Self-learner in reading Machine/Deep Learning Research Paper for better understanding. Driven by curiosity to improve skills & know latest trends.
Mentorship and Guidance
Mentoring ML, dev & version control, empowering with guidance. Inspiring enthusiasts, sharing knowledge at Medium & Live Class.
Featured Projects
Projects
Meeting Intelligence
Meridian — AI Meeting Intelligence
AI meeting intelligence platform that turns recordings into searchable transcripts, structured summaries, and contextual Q&A. Groq for speech-to-text, Google Gemini for summarization and chat, with FastAPI + React/TypeScript, JWT auth, and project-level organization.
FraudNet
Fraud detection on ~15,000 unlabeled signups from an unknown platform running a signup offer. Fellegi–Sunter weighted entity linkage across email aliases, device hash, card, and address produced 251 suspected actor clusters — visualized so you can inspect why accounts were linked.
YouTube
LLM YouTube Metadata Connector
MCP server that exposes YouTube Data API v3 tools to Claude Desktop for tool calling — fetch video data, generate SOP-compliant metadata, and publish updates through a two-step confirmation flow. Used to update 500+ YouTube videos. Prompt engineering, deterministic validation, human-in-the-loop review.
Phishing Domain Detection
ML pipeline to classify URLs as legitimate or malicious using a Random Forest trained on URL, domain, page, and content features. Served via Flask REST API with Cassandra logging every prediction. Deployed to Azure during iNeuron ML internship.
Mixed-Effects Marketing Modeling
Hierarchical mixed-effects regression (LME) to quantify marketing campaign elasticity. Decomposed global vs. channel-level effects to support data-driven budget allocation and ROI optimisation using Marketing Mix Modeling.
Contributing to Major Python Ecosystem Projects
Open Source Contributions
Librosa
Documentation improvement fixing broken SciPy conference paper links
Awaiting stable SciPy proceedings URL
Writing on AI, ML & System Design
Blog
In a Minute
Dense, visual breakdowns of complex ML concepts — Transformers, SVMs, CNNs, Forecasting, Bayesian Learning & more. Built to understand the big picture in a single read.
Web Development
Practical web development — HTTP methods, CORS, lazy loading for UX, and Flask session-based authentication. All with working code examples.
Python
Python fundamentals, tools, and data science implementations — virtual environments, packaging, matrix operations, and performance deep dives.
Latest articles
- How I Built My Data Science Portfolio with Next.js and GitHub Pages (Free)A step-by-step guide to building a free data science portfolio with Next.js static export, Tailwind CSS, and GitHub Pages — plus what to put in it so recruiters actually care.
- Confidence Interval vs Prediction Interval ExplainedAn interactive simulation showing why CI and PI are fundamentally different things. One band covers the mean line. The other catches new individual points.
- Transformer Architecture Explained in a MinuteTransformer architecture explained in 5 minutes: 6 components, Attention formula, encoder vs decoder, GPT vs BERT. Visual breakdown, no fluff.
- Forecasting in a MinuteWhat is forecasting, why it matters, 6 types, and how inline forecasting works — with a real interest-rate decision model walkthrough.
- SVMs & Kernel Trick ExplainedSupport Vector Machines: optimal hyperplane, support vectors, kernel trick, and 4 kernel types with gamma tuning.
- List vs Set in Python — Performance, Use CasesList vs Set in Python: iteration speed, O(1) membership testing, memory trade-offs, and exactly when to use each — with benchmark results.
- Scalars, Vectors, Matrices & Tensors ExplainedFrom a single number to n-dimensional arrays — scalars, vectors, matrices, and tensors defined, notated, and compared in one read.
- CNNs Explained — Convolutional Neural Networks in a MinuteCNN architecture in one read: kernel, stride, padding, pooling, flattening — how images flow from pixels to predictions.
Do You Have Any Questions?
Contact Me
I'M AT YOUR SERVICES
CONNECT WITH ME
I RESPOND WITHIN 24 HOURS