Ritik Kumar Mukhiya
Analyst
Python
SQL
Node JS
JavaScript
Java
Data Pipeline & Feature Engineering: 4/10
Active 9 hours ago
+91 (9570) 467417
Invite to interview
Message
Download CVCV
Overview
Technical skills
Timeline
Roles
Overview
Fullstack web developer focused on static frontend UIs and straightforward Node.js authentication backends at a solid junior-to-middle level. The strongest proven skill is implementing secure authentication flows backed by bcrypt, JWT and Sequelize/Sequelize-adjacent DB wiring as shown in backend/routes/authRoutes.js, middleware/authMiddleware.js and config/db.js. There is little public evidence of automated tests, CI, advanced state machines or measured performance and accessibility work.
Technical skills
Python
SQL
Node JS• Middle
JavaScript• Middle
Java• Junior
Node JS
Express
Databases
Microsoft Fabric
MySQL
Oracle
PostgreSQL
Snowflake
AI/ML
AI/ML
Pandas
NumPy
Scikit-learn
NLTK
Sentiment Analysis
Time Series Forecasting
Classic ML
Statsmodels
DevOps
Azure
GitHub
Analytics
ETL/ELT
Power BI
Matplotlib
Seaborn
Plotly
Timeline
Data Analyst / Power BI Developer
•
Middle
Tata Consultancy Services
•
Full-Time
Developed end-to-end financial analytics for WIP cost tracking using Power BI with DAX, SQL, and Python, automating calculations for faster finance reporting. Built KPI monitoring dashboards to track operational and channel performance and support quicker corrective actions. Created ETL and governed star-schema data models using Azure Data Factory and Oracle Data Integrator, structuring data for reliable enterprise analytics. Migrated Excel-based reporting to automated Power BI workflows to improve accuracy and refresh frequency.
Power BI
SQL
Python
Azure
PostgreSQL
MySQL
Oracle
Lalit Narayan Mithila University
Bachelor's Degree •
Mathematics
Middle AI/ML Engineer
Confidence: Medium Data-centric
Data-centric ML practitioner at a junior-to-early-middle level whose main strength is exploratory data analysis and applied prototyping for sentiment classification and time series forecasting. The strongest proven skill is hands-on EDA and applied forecasting demonstrated by Forecasting.ipynb (date handling, resampling, ExponentialSmoothing, train/test MAPE) and sentiment text cleaning plus polarity/subjectivity extraction in Seniment.ipynb. There is little to no evidence of production MLOps, experiment tracking, automated tests, or custom model development in public code.
Model Architecture & Training
2/10
How well models are designed and trained
Minimal model-building work; uses off-the-shelf sentiment tools (VADER, TextBlob) and classical forecasting (ExponentialSmoothing, LinearRegression) but no custom architectures, training loops, or optimizer/scheduler choices.
Evidence
sentiment_analyis_amazon_review/Seniment.ipynb: application of SentimentIntensityAnalyzer() and TextBlob(Text).sentiment to compute polarity/subjectivity
Time-forecasting-on-gold-price/Forecasting.ipynb: use of ExponentialSmoothing and LinearRegression for forecasting
Data Pipeline & Feature Engineering
4/10
How data is prepared for models
Clear, hands-on data preparation and EDA work including cleaning text, missing-value analysis, datetime parsing, resampling and grouping; typical of data-centric projects but not production-grade pipelines.
Evidence
sentiment_analyis_amazon_review/Seniment.ipynb: missing_value_analysis(), check_dataframe(), text cleaning with regex and mapping reviewText to lowercase
Time-forecasting-on-gold-price/Forecasting.ipynb: building month datetime index, resample('A'/'Q'), grouping and train/test split
Experimentation & Evaluation
3/10
How results are measured and tested
Basic evaluation and experimentation: train/test split, MAPE metric, simple model comparison and confidence-interval style forecast reporting, but no experiment tracking, ablation studies or reproducibility scaffolding.
Evidence
Time-forecasting-on-gold-price/Forecasting.ipynb: mape(actual,pred) function and results DataFrame comparing 'RegressionOnTime' and ExponentialSmoothing
Time-forecasting-on-gold-price/Forecasting.ipynb: pred_df with prediction and +/- 1.96 * std(final_model.resid) used to show forecast intervals
MLOps & Deployment
How models are shipped to production
Not evidenced in public code
Computational Efficiency
How efficiently computing resources are used
Not evidenced in public code
Research Depth & Innovation
Depth of research and new ideas
Not evidenced in public code
Expertise
E-commerce & Retail AI• Junior
Finance & FinTech AI• Junior
Industries
Commerce• Junior
Financial Services• Junior
Technologies
AI/ML
Classic ML
Python
SQL
MySQL
PostgreSQL
Snowflake
Oracle
Azure
Time Series Forecasting
Sentiment Analysis
GitHub
Microsoft Fabric
Recommendations
- Lead rapid prototyping and EDA projects that convert CSV/structured data into analysis-ready features and visualizations for stakeholders.
- Develop small-scale forecasting or sentiment prototypes that require clear train/test evaluation and simple baseline comparisons.
- Build dashboards or analyst-facing notebooks that package the existing EDA and forecast outputs into repeatable reports.
- Focus on maturing pipelines by adding unit tests, reproducible preprocessing scripts, and basic model serialization to prepare prototypes for production handoff.
Repositories
The developer's experience in this domain has been verified based on AI analysis of the following repositories:
Middle Data Scientist
Confidence: Low ML Practitioner
An early-career ML practitioner focused on applied predictive modeling and exploratory data analysis with practical, notebook-first implementations. The strongest proven skill is applied time-series forecasting and baseline comparison, shown by building a regression baseline, ExponentialSmoothing forecasts, simple confidence intervals, and MAPE-based evaluation. There is limited evidence of production readiness, rigorous statistical validation, reproducible environments, or scalable data engineering in public artifacts.
Statistical Rigor
2/10
Correct use of statistics
Basic model evaluation is present (MAPE, simple CI from residuals) but there are no formal assumption checks, no stationarity tests for time series, limited uncertainty quantification, and no rigorous statistical validation.
Evidence
Time-forecasting-on-gold-price/Forecasting.ipynb: mape function and MAPE printed for models
Time-forecasting-on-gold-price/Forecasting.ipynb: prediction CI computed as prediction +/- 1.96 * std(final_model.resid)
Data Wrangling & Cleaning
4/10
Preparing and cleaning data
Reasonable ad-hoc cleaning and basic data-quality checks are implemented (missing-value analysis, text cleaning, column drops), but pipelineization, provenance, and leakage-prevention controls are missing.
Evidence
sentiment_analyis_amazon_review/Seniment.ipynb: missing_value_analysis and check_dataframe functions
sentiment_analyis_amazon_review/Seniment.ipynb: reviewText cleaning via rt = lambda x: re.sub("[^a-zA-Z]", ' ', str(x)) and lowercasing
Exploratory Analysis & Visualization
3/10
Exploring and visualizing data
Multiple visualizations are produced and labelled, but interpretations and written insights are sparse; plots are used more as exploratory outputs than as a structured data-story with hypothesis-driven commentary.
Evidence
Exploratry-Gym-Data-Analysis/Gym.ipynb: many Plotly and matplotlib visualizations (pie charts, bar plots, grouped comparisons)
Time-forecasting-on-gold-price/Forecasting.ipynb: month_plot and multiple resampling/time-aggregation plots
Predictive Modeling
3/10
Building models that predict
Predictive work shows baseline-first thinking and model comparison (linear regression on time then ExponentialSmoothing) and a simple evaluation metric, but lacks proper cross-validation, hyperparameter search, deeper error analysis, calibration or robustness checks.
Evidence
Time-forecasting-on-gold-price/Forecasting.ipynb: baseline LinearRegression on time followed by ExponentialSmoothing final_model
sentiment_analyis_amazon_review/Seniment.ipynb: rule-based sentiment labeling using VADER scores and TextBlob polarity/subjectivity
Business Insight & Impact
1/10
Turning analysis into business value
Little to no business framing or cost-sensitive reasoning is present; analyses produce visualizations and metrics but do not link results to business KPIs or error-cost tradeoffs.
Evidence
Time-forecasting-on-gold-price/Forecasting.ipynb: descriptive printouts and plots of gold price but no business-impact or cost-of-error discussion
Reproducibility & Notebook Hygiene
1/10
Clean, repeatable analysis
Notebooks lack pinned environments, reproducible dependency management, tests, seeds, or exportable pipeline code. There are in-notebook pip installs and ad-hoc cells rather than structured, reproducible workflows.
Evidence
sentiment_analyis_amazon_review/Seniment.ipynb: in-notebook pip install commands and mixed/mistyped pip outputs
NOTEBOOK META entries: low reasoning_ratio and iterative execution style across notebooks
Industries
Commerce• Middle
Financial Services• Middle
Lifestyle• Junior
Technologies
Scikit-learn
Seaborn
Matplotlib
Plotly
Pandas
NumPy
NLTK
Statsmodels
Recommendations
- Develop small to medium PoCs that require time-series forecasting or quick sentiment prototypes for product teams, focusing on clear baselines and simple evaluation metrics.
- Work on cleaning and modularizing notebooks into reproducible scripts or small pipelines (requirements.txt/conda env, fixed random seeds, functions instead of monolithic cells).
- Collaborate on projects that convert exploratory analyses into tested, deployable components (model evaluation pipelines, CI, and basic monitoring).
- Practice statistical rigor by adding assumption checks, stationarity tests, cross-validation, and written interpretations for each key finding.
Repositories
The developer's experience in this domain has been verified based on AI analysis of the following repositories:
Middle Frontend Developer
Confidence: Medium Fullstack
Fullstack web developer focused on static frontend UIs and straightforward Node.js authentication backends at a solid junior-to-middle level. The strongest proven skill is implementing secure authentication flows backed by bcrypt, JWT and Sequelize/Sequelize-adjacent DB wiring as shown in backend/routes/authRoutes.js, middleware/authMiddleware.js and config/db.js. There is little public evidence of automated tests, CI, advanced state machines or measured performance and accessibility work.
UI Component Architecture
2/10
How interface parts are built
Component architecture is minimal and file-based; mostly static HTML pages and global CSS rather than composable components or a design-system authored by the developer.
Responsive & Cross-browser
3/10
Works on all screens and browsers
Responsive work uses classic media queries and grid layouts and covers common breakpoints, but lacks modern fluid/layout features like container queries, explicit RTL/i18n support or feature-detection fallbacks.
Performance Optimization
2/10
Speed of the interface
Small signs of performance awareness (font preconnect) but no measured optimizations, code-splitting, lazy-loading, or tooling artifacts (bundle reports, analyzer traces).
Accessibility & Semantics
2/10
Usable for everyone
Basic semantic HTML is present, but accessibility work is limited: images lack alt attributes, interactive elements lack labels/ARIA and there are no CI a11y checks or explicit focus management patterns.
State Management & Data Flow
3/10
Managing data in the app
Server-side state and auth are implemented coherently with hashed passwords, JWT generation and middleware verification, but there is no evidence of advanced server-state discipline like cache invalidation, request cancellation, optimistic updates or test coverage.
Evidence
rmUKHIYA/auth-api/backend/routes/authRoutes.js: signup and signin flows using bcrypt and jwt with error handling
rmUKHIYA/auth-api/backend/middleware/authMiddleware.js: JWT verification middleware setting req.user
rmUKHIYA/auth-api/backend/server.js: sequelize DB connection and sync, express JSON and CORS setup
UX & Visual Polish
3/10
Look and feel quality
UI shows considered visual polish and UX details such as hero overlay, hover transitions and grid layouts, but lacks loading/error states, skeletons, undo flows or perceptual performance techniques.
Expertise
HTML & CSS• Middle
Industries
Education• Middle
Travel & Tourism• Junior
Technologies
JavaScript• Middle
Node JS• Middle
Express
Recommendations
- Convert repeated UI blocks into components (React/Svelte/Stencil) and introduce a design-system or token-based CSS approach to improve reusability and testing.
- Add automated tests for backend auth flows and a minimal CI pipeline to run linting and tests before deploys.
- Improve frontend UX resilience by adding explicit loading, error and empty states, plus optimized image loading (lazy, srcset) and measured performance checks with Lighthouse.
- Address accessibility gaps: add alt text, form labels, keyboard focus management and run axe/ESLint-a11y in CI.
Repositories
The developer's experience in this domain has been verified based on AI analysis of the following repositories: