SQL
Joins, CTEs and window functions that turn raw tables into the numbers a business runs on.
Proof · Retail sales analysisWrote the SQL behind revenue, category and seasonal KPIs on 25,000+ sales records.JoinsCTEsWindow functions
Skills
What I can do, with proof
Every skill links to the project or job where I used it.
Power BI
Data models, DAX measures and dashboards people actually open, and that refresh on their own.
Proof · Retail sales analysisAutomated the dashboard refresh and cut manual reporting effort by 60%.DAXData modellingScheduled refresh
Excel
Pivot tables, lookups and Power Query for fast cleaning, checks and reporting.
Proof · HR analyst internshipCleaned recruitment data and built hiring KPI dashboards across departments.Pivot tablesLookupsPower Query
Python
pandas and NumPy to clean and analyse data, Matplotlib and Seaborn to show it.
Proof · ETL pipelineBuilt the cleaning pipeline (deduplication, type checks, missing values) that raised data quality by 32%.pandasNumPyMatplotlibSeaborn
Statistics
Descriptive statistics, regression, hypothesis tests and clear experiment readouts.
Proof · Product pilot at AUDIOEDefined the engagement metrics for a 25-user pilot and analysed the results (about 80% positive).RegressionHypothesis testingCorrelation
Machine learning and AI
scikit-learn models, clustering and LLM features, tested properly before anyone relies on them.
Proof · Alzheimer's predictionBuilt and validated a CNN on MRI scans: 90.2% accuracy, F1 0.89.scikit-learnCNNK-meansLLMs
How I work
Stakeholder communicationRequirements gatheringData storytellingKPI reportingDashboard automationEDA
My work
Projects I've built
Python · SQL · Power BI
25,000+textile sales records
Retail sales analysis
- Built an ETL pipeline that improved data quality by 32% through deduplication, type checks and missing-value handling.
- Wrote the SQL behind KPIs on revenue, category performance and seasonal demand.
- Automated the Power BI dashboard refresh, cutting manual reporting effort by 60%.
PythonSQLPower BIETL
Python · Pydantic · FastAPI
0ways for feed content to change the agent's actions
Warehouse feed agent
- Reads warehouse updates over HTTP (JSON, JSONL or CSV) and commits, rejects or quarantines each one.
- A fixed 8-stage pipeline: sanitise, scan for injected instructions, validate, replay check, reference lookup, business rules, state check.
- Every decision goes into a hash-chained audit log that can be verified later. Runs fully offline; the AI classifier is optional.
PythonPydanticFastAPIpytest
GitHubscikit-learn · React · FastAPI
K-meansclustering + LLM explanations
AI career guidance system
- Grouped users with K-means clustering on psychometric assessment data.
- Added LLM-written explanations on top of each cluster so every user gets a report they can understand.
Pythonscikit-learnReactFastAPI
Python · CNN
90.2%accuracy · F1 0.89
Alzheimer's disease prediction
- Built and validated a convolutional neural network on an MRI dataset.
- Controlled overfitting with augmentation, class balancing and cross-validation.
Pythonscikit-learnCNN
Experience
Where I've worked
Sep 2025 – Dec 2025AUDIOE · edtech startup, WE Hub
Team Lead, Product and Data
- Defined engagement metrics for a 25-user product pilot, then collected and analysed usage and feedback data (about 80% positive engagement).
- Turned the pilot findings into recommendations that shaped the product direction and an accelerator application.
- Coordinated a 10-person team across requirements, timelines and communication.
Apr 2025 – Sep 2025Morepen Proprietary Drug Research
Human Resources Analyst Intern
- Cleaned and analysed recruitment data in Excel and SQL, fixing inconsistencies that were distorting hiring reports.
- Built dashboards tracking hiring KPIs across departments, including R&D, giving managers one view of pipeline and time-to-fill.
- Used generative AI and prompt engineering to speed up the weekly hiring reports.
Jan 2025 – Mar 2025Qriocity · freelance
Data and Analytics Technical Content Writer
- Wrote tutorials on SQL analytics, Python data analysis and statistics, each with reproducible code and outputs.
- Covered data modelling, query optimisation and EDA, and checked every query and result before publishing.
Education
Study
Jan 2026 – presentUniversity of Hertfordshire, UK
MSc Artificial Intelligence and Robotics
Dec 2021 – Apr 2025CMR Technical Campus, Hyderabad
B.Tech Artificial Intelligence and Machine Learning
Certifications
- CERTData Analysis with PythonfreeCodeCamp
- CERTGoogle Cloud Data Analyst TrackGoogle Cloud
- CERTMicrosoft Power BIIIT Madras
- CERTMachine Learning with PythonfreeCodeCamp
Recognition
- 2023Smart India Hackathon, finalistIoT lighting solution
- 2023Technovanza 3.0 awardSmart city innovation
- 2024WE Hub recognitionAUDIOE proof of concept