Hey there, I'm
Data Analyst → Data Scientist — 2+ years at Rockwell Automation (2023–2025) building analytical systems on real industrial data. Now fully focused on ML, AI, and data science.
Who I am, what drives me, and where I'm headed.
To bridge industrial operations and data intelligence — building AI systems that are explainable, fair, and impactful for Industry 4.0 and beyond.
Industry 4.0 · Predictive Maintenance · Manufacturing Analytics · NLP · Explainable AI · Generative AI · Recommendation Systems
All projects are open source under @CodeWithSrish. Working in Jupyter Notebook and Python with Streamlit deployments.
↗ github.com/CodeWithSrishPredictive maintenance ML, industrial time-series analytics, production-ready ML deployment, and LLM-powered systems.
Industrial Analytics · Data Science · Generative AI · Explainable AI · Electronics & Communication — a rare cross-disciplinary combination.
Industry analyst roots, growing into data science — the combination that makes my work different.
Managed and analyzed real-time process data from industrial automation systems to identify operational bottlenecks, surface efficiency gaps, and deliver data-driven recommendations to engineering stakeholders.
Conducted data-driven RCA on automation logs using statistical methods — hypothesis testing, trend analysis — to diagnose failure modes and reduce system downtime, directly informing predictive maintenance strategies.
Developed performance tracking dashboards and reports translating raw engineering metrics into executive-ready insights — enabling faster, evidence-based decisions by cross-functional teams.
Spearheaded end-to-end upgrade of legacy PLC systems to enable high-frequency data extraction — building the pipeline infrastructure required for predictive modeling and laying the groundwork for ML-driven automation.
2+ years at Rockwell Automation gave me something most data scientists don't have: real production data experience. I know what it costs when a system goes down, how industrial sensor data is actually generated, and what "actionable insight" means when an engineer is staring at a live fault. That foundation — statistical RCA, BI reporting, pipeline infrastructure — is now the lens I bring to ML. The transition isn't a career pivot; it's a natural progression. I completed dual certification in Data Science & AI (BIA 2026) while building production-grade projects in fraud detection, credit risk modeling, NLP, and recommendation systems. Every model I build is grounded in the analyst instinct I developed on the job — I don't just know how to run an algorithm, I know why it matters to the business.
Real-world ML & AI — all open source on @CodeWithSrish
End-to-end NLP system for misinformation detection. Advanced preprocessing, feature engineering, model evaluation, explainability, and a Streamlit interface.
↗ View on GitHubXGBoost credit risk prediction with statistical analysis, SHAP & LIME explainability, AIF360 fairness evaluation, and Streamlit deployment.
↗ View on GitHubMovie recommendation system using collaborative filtering and ML, delivering personalised suggestions through an interactive application.
↗ View on GitHubEnd-to-end fraud detection system on 11,142+ real transactions. XGBoost champion model with 100% accuracy, 99.8% F1, ROC-AUC 1.0, SHAP explainability, and ₹282 Cr projected annual savings. Features forensic feature engineering including the "Drain-to-Zero" pattern.
↗ View on GitHubComprehensive expertise across the AI lifecycle, from statistical modeling to generative systems.
Open to internships, collaborations, and conversations about AI & data.
Based in India, building for the world.
Currently pursuing B.Tech at MERI (MDU), New Delhi. Open to remote opportunities globally and on-site roles in Delhi NCR.
28.6139° N, 77.2090° E — Open to opportunities worldwide