About · Curriculum Vitae
Data Analytics Postgraduate, service manager,
behavioural data explorer.
I finished an MSc in Data Analytics at London Metropolitan University in July 2026, moving from mathematics and education into clinical and behavioural analytics. The two undergraduate degrees listed below are often read as a pair of BScs — they aren't: a B.Sc. in Mathematics, then a Bachelor of Education. One taught me to model, the other to explain. The day job taught me to handle data that has to be right the first time.
My dissertation — HF-RISK — audits leakage in heart-failure outcome prediction and still lands at 0.710 AUROC internally. My behavioural work turns 845 messy human trials into a readable signal. Both share one rule: honest numbers only.
London, UK
Location
MSc Data Analytics
Completed Jul 2026
Ladbrokes / Entain
Day role
Open to graduate roles
Availability
01 · Experience & Education
Three strands, one stack.
Service leadership, AI evaluation and analytics research — each strand feeds the others.
Current experience
Customer Service Manager — Ladbrokes / Entain
- Floor leadership — shift rota planning, complaint resolution, team coaching
- Service standards — coaching staff under Gala Coral Group brand guidelines
- Rota management — staffing allocation across rotating shifts
- Customer escalation — flags, resolves, and follows through on issues
Education
- MSc Data Analytics — London Metropolitan University Jul 2026
- B.Sc. Mathematics — Gujarat University Jun 2021
- Bachelor of Education — Gujarat University Apr 2023
02 · Timeline
By the numbers.
The full record — every role, every project, every date I can stand behind.
- Apr 2026 — Leadership
Customer Service Manager — Ladbrokes / Entain, Harrow
Leading the floor: shift rota, complaint resolution, and coaching a team on service standards under Gala Coral Group brand guidelines.
- Nov 2025 – May 2026 LLM evaluation
AI Response Evaluator — TMX China
Scored and calibrated AI-generated responses against rubric quality bars; fed systematic failure patterns back to the model team.
- Oct 2025 – present 845 trials
Card-sorting analysis — internship under Dr. Subeksha Shrestha
Behavioural data internship (London Met module): cleaned and analysed 845 recorded trials of spatial card-sorting; built the interactive analysis viewer.
- Sep 2025 – Jul 2026 MSc · Jul 2026
MSc Data Analytics — London Metropolitan University
Completed July 2026. Dissertation: HF-RISK — clinical outcome prediction with a full leakage audit, a patient-grouped MIMIC-IV extension, and a Streamlit prototype.
- May 2025 – Apr 2026 Compliance
Admin — PHC Home Care
Care records, log validation, incident tracking and audit readiness; built the website for a secondary company (chi ezahomecare.co.uk).
- Aug 2025 Community
Community Volunteer — Metropolitan Police
Safer Neighbourhoods volunteering: community engagement and local safety outreach in London.
- Feb – Mar 2026 +0.0138 AUROC
HF-RISK dissertation — heart failure outcome prediction
Four horizons (28-day, 3-month, 6-month mortality; 6-month readmission) on 2,008 Zhang/PhysioNet patients, reduced to 143 predictors by a full leakage audit. XGBoost: 0.710 AUROC internal at six months; 0.599 on MIMIC-IV without retraining. Random Forest wins the short horizons (0.781 / 0.839). Streamlit prototype reviewed with UK cardiologists. Extended later in 2026: a patient-grouped MIMIC-IV retrain showed GCS components add +0.0138 AUROC at 28 days beyond a 34-feature baseline, and still add +0.0104 on top of an ICU-stay control.
- Prior B.Sc. + B.Ed.
B.Sc. Mathematics (Jun 2021) · Bachelor of Education (Apr 2023) — Gujarat University
Two separate undergraduate degrees, not two of the same: mathematics first, then a Bachelor of Education. One taught the modelling, the other the explaining.
03 · The toolbelt
Languages & methods.
What I reach for, in order of frequency.
Languages & libraries
- Python
- pandas
- NumPy
- scikit-learn
- SHAP
- Plotly
- Streamlit
- lifelines
- matplotlib
- SQL
- R
- JavaScript
Methods
- Statistical Modelling
- Predictive Analytics
- Data Cleaning
- Feature Engineering
- Data Visualisation
- Power BI
- Behavioural Analysis
- Compliance-aware data handling
04 · What I'm like on a team
Professional, curious, careful.
Professional skills
- Attention to detail
- Data integrity & compliance
- Analytical thinking
- Clear communication
- Process improvement
- Stakeholder collaboration
- Professionalism under pressure
- Confidential data handling
Beyond work
Ancient history
Especially Egypt — temple economies to tomb logistics.
Fitness & running
Long slow distance; discipline as data collection.
Self-development
Habits, systems, and honest self-review.
Continuous learning
New methods, new datasets, new questions.
How I work
- Checked before shown
- Every figure on this site can be traced to a script and a run log. If a number can't be reproduced, it doesn't get published.
- Empty means empty
- I don't impute a clinical value to make a table look complete, and I don't drop an inconvenient subgroup to tidy a result.
- Limits travel with the result
- Sample size, cohort selection and dataset shift are written next to the number, not buried in a footnote.