Hi there, I’m Yusuf Tiamiyu and I currently work at Cerba Lancet Nigeria where lab data connects to real patient outcomes. I’ve spent the last two years turning messy clinical numbers into decisions that actually get made.
I’m a data analyst at Cerba Lancet Nigeria, one of West Africa’s largest diagnostic lab networks. I hold two roles there: general data analyst and Data Protection Officer. Most days it’s a bit of both.
The work I do sits at the intersection of healthcare and numbers that carry real weight. Turnaround times on lab tests. Trends in operational data. Supply chains for reagents departments cannot run out of. Patterns in clinical data that, if you spot them early, change how care gets delivered.
I got into analytics because I always needed to understand why. Why is this process slow? Why do we keep ordering too much of one thing and running out of another? Data gives you something better than intuition. It gives you evidence. And once you have that, you stop guessing.
Outside work I’m a gamer — which probably explains why I treat every dataset like there’s something buried in it worth finding. There usually is. Co-author of a peer-reviewed paper in Hydrology Journal (2023).
Rebuilt from the ground up: React/Supabase platform replacing an earlier MySQL/Flask version. Row-level filtering across site, department and TAT type, security hardening, and analytics (breach Pareto, shift×site heatmap) built to answer specific operational questions. In daily clinical use across 9 sites.
Excel dashboard monitoring test volumes, TAT %, staff adequacy across sites. Dynamic filtering, auto direction arrows.
Excel system for 100+ lab supplies. Auto-flags REORDER NOW at two-month buffer. Shifted procurement from reactive to planned.
SQL deep-dive into global COVID death and vaccination data. Death rates, infection vs population, continental breakdowns, rolling totals.
Four-phase MySQL pipeline. Deduplication via ROW_NUMBER, standardisation, intelligent null filling with self-joins, removal of dead rows.
Exploratory analysis on the cleaned dataset. Companies that went to zero, rankings with DENSE_RANK, rolling cumulative totals.
Chains 12 advanced SQL concepts in sequence. Recursive CTEs for date generation, LAG/LEAD for intelligent null filling. Started from real frustration with UNION ALL.
Excel dashboard on 1,000 customer records. KPI cards, five purchase-rate analyses, slicers for four dimensions. Middle-age group and 2–5 mile commuters drove the highest conversion rates.
Tableau dashboard exploring 3,818 listings. Bedroom demand, pricing by location, monthly revenue seasonality, room type split. Joined Listings and Calendar datasets in Tableau Public.
Every dataset is noise until you know what you are actually trying to answer. I spend real time here, not just on what is being asked, but on what decision is waiting on the other side of it.
Duplicates, nulls, wrong data types, inconsistent formatting. I treat every raw dataset like it is lying until I can prove otherwise. Clean data first. Analysis second. Every single time.
I go in with questions and come out with more. I keep slicing by time, region, category and rank until the data stops giving me anything new. That is when the EDA is done.
A finding nobody can act on is not a finding. Everything I deliver has a decision attached. A reorder threshold. A staffing flag. A process that needs to change. Insight without action is just trivia.
Open to data analyst roles, clinical research projects, and anything that involves making sense of messy data. Lagos-based, remote-friendly.