Engineered automated career recommendations for 500 students by implementing a Text-to-SQL agent using Lang Graph and FastAPI. Reduced agent errors by 80% and cut query generation time by 60% by redesigning SQL views to flatten complex relational data. Achieved 95% precision in fraud detection by implementing a real-time system using Benford's Law, Isolation Forest, and text anomaly detection. Reduced false positives by 75% by refining rule-based logic for duplicate transactions and unusual timestamp patterns.

Software Developer Intern at IntelliSOFT Consulting Ltd

July 2, 2024 - September 27, 2024

Accelerated 3 project timelines by 20% by developing FHIR Implementation Guides and HL7 artifacts (FSH) Improved data interoperability for a 5-developer team by creating standardized OpenMRS data dictionaries. Optimised developer onboarding for OpenMRS by 1 month for 1 developer, speeding up development work

Education

Bachelor of Science in Computer Science at The University of Nairobi

September 1, 2021 - September 1, 2025

Qualifications

Add your qualifications or awards here.

Industry Experience

Software & Internet, Retail, Healthcare, Transportation & Logistics, Professional Services

Diamond Price Predictor

The Problem: Most Buyers Overpay — Even on Small Diamonds

You’re shopping for a small to mid-sized diamond (0.2–1.5 carat) — maybe for an engagement ring, gift, or investment.
But jewelers exploit confusion:

Same 0.7ct diamond? $1,500 at one store, $3,200 at another
No transparency on carat-to-price curve
“Premium cut” or “rare color” excuses hide 100%+ markups
Result: Everyday buyers overpay $500–$2,000 on modest stones.

The Solution: Instant, Data-Backed Price Estimates

No jargon. No GIA reports. Just one number.

Enter carat weight → Get real market price range in 1 second.

Data Scientist

Jumbo Global E-Commerce Operations Dashboard

Jumbo is a sample global e-commerce company. The company has data on sales that is underutilised after transactional processing. This project analyses sales data for the Head of Global Operations in order to improve global operations.

Achievements:

Protected margins for the Head of Global Operations by diagnosing a 12% return rate in Fashion goods via the Streamlit dashboard
Guided holiday strategy for Marketing Executives by uncovering a 20% seasonal revenue surge through SQL time-series trend analysis
Unlocked logistics savings for the Supply Chain department by proving zero correlation between delivery speed and CSAT via Python statistical analysis.

Tools: