const { useState, useEffect, useRef, useCallback } = React; // ─── DATA ──────────────────────────────────────────────────────────────────── const PROJECTS = { "ML Apps": [ { title: "Rock Classifier App", desc: "A deep learning system that identifies 41 rock types from a single field photograph, using a ConvNeXt-Tiny backbone with a custom dual-head architecture that predicts rock group (igneous / metamorphic / sedimentary) and specific rock type at once. It's a fully productionized system, not just a model: a FastAPI backend serving real-time inference, a React frontend, and a closed feedback loop where users can correct a wrong prediction — each correction is persisted to a managed database as future retraining data. When the original hosting provider unexpectedly took the backend offline, I migrated the entire service — inference backend and feedback pipeline alike — to a new cloud stack with zero data loss, redesigning it to scale to zero instead of running (and billing) around the clock.", stack: ["PyTorch","FastAPI","React","Google Cloud Run","Supabase","GitHub Actions CI/CD"], live: "https://fehintiti.github.io/Rock-Classifier-App/", github: "https://github.com/Fehintiti/Rock-Classifier-App", featured: true, img: window.ROCK_APP_SCREENSHOT, }, { title: "Multiple Disease Prediction", desc: "Clinical ML app classifying Heart Disease, Parkinson's, and Diabetes from patient input data. Deployed as an accessible Streamlit application.", stack: ["Scikit-learn","Streamlit","Python","Pandas"], github: "https://github.com/Fehintiti/Multiple-Disease-Prediction-ML-App", img: "https://images.unsplash.com/photo-1576091160550-2173dba999ef?w=600&q=80", }, { title: "Zenodo Product Classification", desc: "NLP pipeline for automated product categorisation using TF-IDF, Word2Vec embeddings, SVM, XGBoost and neural networks with hyperparameter tuning.", stack: ["NLP","Text Classification","Python","Scikit-learn"], github: "https://github.com/Fehintiti/Zenodo-Product-Classification", img: "https://images.unsplash.com/photo-1555949963-ff9fe0c870eb?w=600&q=80", }, ], "Cloud Pipelines": [ { title: "Gold Price Intelligence Pipeline", desc: "Fully serverless data engineering pipeline ingesting, processing and serving live precious metals data. EventBridge triggers Lambda, raw files land in S3, Athena queries transform the data.", stack: ["EventBridge","Lambda","S3","Athena","JavaScript"], live: "https://fehintiti.github.io/gold-pipeline-dashboard/", github: "https://github.com/Fehintiti/gold-pipeline-dashboard", featured: true, img: "https://images.unsplash.com/photo-1611974789855-9c2a0a7236a3?w=1200&q=80", }, { title: "Mining Operation Optimisation", desc: "Predictive maintenance model for industrial mining equipment. Processes sensor data to forecast failure events and fuel consumption, enabling proactive scheduling.", stack: ["Python","Predictive Maintenance","ML","Pandas"], github: "https://github.com/Fehintiti/Mining-Operation-Optimization-App-Project", img: "https://images.unsplash.com/photo-1504307651254-35680f356dfd?w=600&q=80", }, { title: "Equipment Failure Prediction", desc: "Sensor-based failure prediction for industrial machinery. Time-series readings detect early failure signals, enabling maintenance teams to act before breakdowns occur.", stack: ["Python","Time Series","Classification","Scikit-learn"], github: "https://github.com/Fehintiti/Mining-Equipment-Failure-", img: "https://images.unsplash.com/photo-1581091226825-a6a2a5aee158?w=600&q=80", }, ], "BI & Analytics": [ { title: "Southern Water UK Dashboard", desc: "Full BI project processing 89,000+ operational records. SQL data pipeline feeds a Power BI dashboard surfacing key operational KPIs for non-technical stakeholders.", stack: ["SQL","Power BI","Data Modelling","ETL"], featured: true, linkedin: "https://www.linkedin.com/posts/activity-7434568295699099648-snUL?utm_source=share&utm_medium=member_desktop&rcm=ACoAABVw1KQBpKhZRw5AUfhVAMLuBpLNx6Ym3Ao", img: "https://images.unsplash.com/photo-1460925895917-afdab827c52f?w=1200&q=80", }, { title: "NYC Green Taxi Challenge", desc: "Maven Analytics challenge entry — four years of NYC Green Taxi records analysed in SQL and visualised in Power BI with actionable pricing and demand recommendations.", stack: ["SQL","Power BI","Maven Challenge","Data Analysis"], github: "https://github.com/Fehintiti/NYC-Green-Taxi", img: "https://images.unsplash.com/photo-1496442226666-8d4d0e62e6e9?w=600&q=80", }, { title: "Gold Price Analysis 1978–2021", desc: "Longitudinal analysis of 43 years of precious metals data identifying macro trends across inflation cycles, financial crises, and commodity booms.", stack: ["Python","Pandas","Trend Analysis","Finance Data"], github: "https://github.com/Fehintiti/Gold_Price-Analysis-", img: "https://images.unsplash.com/photo-1618044733300-9472054094ee?w=600&q=80", }, ], "Computer Vision": [ { title: "Quartz Core Object Detection", desc: "YOLOv8 model trained to automatically identify and localise quartz minerals in diamond drill core images. Trained via Roboflow with full case study published on Medium.", stack: ["YOLOv8","Roboflow","Object Detection","Python"], github: "https://github.com/Fehintiti/Core-Images-Quartz-Object-Detection", medium: "https://medium.com/@Fehintiti/quartz-core-images-object-detection-af15b1c6c3fa", featured: true, img: "https://images.unsplash.com/photo-1507413245164-6160d8298b31?w=1200&q=80", }, ], "Data Wrangling": [ { title: "Nashville Housing Data Cleaning", desc: "SQL-only data quality project transforming raw Nashville housing records into a fully analysis-ready dataset — address standardisation, null handling, deduplication.", stack: ["SQL","Data Cleaning","ETL","Data Quality"], github: "https://github.com/Fehintiti/Nahville-Housing-Dataset-Cleaning", featured: true, img: "https://images.unsplash.com/photo-1558494949-ef010cbdcc31?w=1200&q=80", }, { title: "Cyclist Bike Share Capstone", desc: "Google Data Analytics certificate capstone. Full analysis of Cyclistic bike-share usage patterns to identify behavioural differences and inform membership conversion strategy.", stack: ["SQL","R","Google Capstone","Data Analysis"], github: "https://github.com/Fehintiti/Cyclist-Bike-Share", img: "https://images.unsplash.com/photo-1534787238916-9ba6764efd4f?w=600&q=80", }, { title: "Covid-19 Data Exploration", desc: "SQL-driven exploration of global Covid-19 infection, mortality, and vaccination data. Tracks rolling case rates across countries and documents vaccination rollout effects.", stack: ["SQL","Python","Exploratory Analysis","Public Health"], github: "https://github.com/Fehintiti/Covid-Project", img: "https://images.unsplash.com/photo-1584036561566-baf8f5f1b144?w=600&q=80", }, ], }; const SKILLS = [ "Python","SQL","Power BI","PyTorch","FastAPI","React", "Scikit-learn","Streamlit","Pandas","Git","Docker", "Jupyter","Matplotlib","NLP","TensorFlow","Tableau","Spark","dbt", ]; const STATIC_ARTICLES = [ { title: "Can AI Really Identify Rocks? A Data Scientist's Challenge", date: "Feb 2026", desc: "A 5-phase ML experiment testing whether computer vision can replicate geological expertise — and what the results reveal about the real limits of AI.", url: "https://medium.com/@Fehintiti/can-ai-really-identify-rocks-a-data-scientists-challenge-347cde82e203" }, { title: "I Bought Gold. Then It Crashed. So I Built a Pipeline to Track It.", date: "Apr 2026", desc: "How a personal investment decision became a full serverless data engineering system with automated ingestion, cloud storage, SQL analytics, and a live dashboard.", url: "https://medium.com/@Fehintiti/i-bought-gold-then-it-crashed-so-i-built-a-pipeline-to-track-it-039211d306db" }, { title: "Git and GitHub: The Data Scientist's Best Friend", date: "Aug 2025", desc: "A practical guide to version control that transforms how data scientists manage, share, and showcase their work across projects and collaborators.", url: "https://medium.com/@Fehintiti/git-and-github-the-data-scientists-best-friend-you-didn-t-know-you-needed-6101d9032f42" }, { title: "Quartz Core Images: Object Detection", date: "Aug 2024", desc: "A full walkthrough of training YOLOv8 to detect quartz minerals in diamond drill core photographs, from Roboflow dataset preparation through to results.", url: "https://medium.com/@Fehintiti/quartz-core-images-object-detection-af15b1c6c3fa" }, { title: "Multi-Element Gold Prediction with Machine Learning", date: "Apr 2023", desc: "Building and deploying a Random Forest model to predict gold concentration from geochemical assay data, with a Streamlit interface for non-technical users.", url: "https://medium.com/@Fehintiti/multi-element-gold-prediction-with-machine-learning-f7608d7bbaa" }, { title: "Exploratory Data Analysis on Mining Geochemical Data", date: "Jun 2023", desc: "Step-by-step EDA across three mineral survey zones covering data cleaning, QA/QC, anomaly detection, and multi-element correlation analysis.", url: "https://medium.com/@Fehintiti/exploratory-data-analysis-eda-on-mining-geochemical-data-4d6657c16058" }, ]; // ─── ICONS ───────────────────────────────────────────────────────────────── const GithubIcon = () => ; const LinkedInIcon = () => ; const MediumIcon = () => ; const TwitterIcon = () => ; const EmailIcon = () => ; const ArrowIcon = () => ; // ─── COUNTER ───────────────────────────────────────────────────────────────── function useCounter(target, duration, triggered) { const [count, setCount] = useState(0); useEffect(() => { if (!triggered) return; let start = 0; const numTarget = parseInt(target.replace(/\D/g, "")); const step = Math.ceil(numTarget / (duration / 16)); const timer = setInterval(() => { start += step; if (start >= numTarget) { setCount(numTarget); clearInterval(timer); } else setCount(start); }, 16); return () => clearInterval(timer); }, [triggered]); return count; } // ─── SKILL BUBBLES ──────────────────────────────────────────────────────────── function SkillBubbles() { const canvasRef = useRef(null); const bubblesRef = useRef([]); const animRef = useRef(null); const activeRef = useRef(true); const initBubbles = useCallback((canvas) => { const tones = ["#222222","#333333","#444444","#555555","#666666","#7B7B7B","#888888","#999999"]; bubblesRef.current = SKILLS.map((skill, i) => { const radius = Math.max(40, Math.min(62, skill.length * 5 + 32)); return { x: radius + Math.random() * (canvas.width - radius * 2), y: radius + Math.random() * (canvas.height - radius * 2), vx: (Math.random() - 0.5) * 1.2, vy: (Math.random() - 0.5) * 1.2, radius, label: skill, color: tones[i % tones.length], opacity: 0, }; }); }, []); useEffect(() => { const canvas = canvasRef.current; if (!canvas) return; const ctx = canvas.getContext("2d"); const resize = () => { canvas.width = canvas.offsetWidth; canvas.height = canvas.offsetHeight; initBubbles(canvas); }; resize(); window.addEventListener("resize", resize); const handleVisibility = () => { if (!document.hidden) { activeRef.current = true; if (!animRef.current) animate(); } else activeRef.current = false; }; document.addEventListener("visibilitychange", handleVisibility); const animate = () => { if (!activeRef.current) { animRef.current = null; return; } ctx.clearRect(0, 0, canvas.width, canvas.height); bubblesRef.current.forEach((b, i) => { b.opacity = Math.min(1, b.opacity + 0.018); b.x += b.vx; b.y += b.vy; if (b.x - b.radius < 0) { b.x = b.radius; b.vx = Math.abs(b.vx); } if (b.x + b.radius > canvas.width) { b.x = canvas.width - b.radius; b.vx = -Math.abs(b.vx); } if (b.y - b.radius < 0) { b.y = b.radius; b.vy = Math.abs(b.vy); } if (b.y + b.radius > canvas.height) { b.y = canvas.height - b.radius; b.vy = -Math.abs(b.vy); } bubblesRef.current.forEach((b2, j) => { if (i >= j) return; const dx = b2.x - b.x, dy = b2.y - b.y; const dist = Math.sqrt(dx*dx+dy*dy); const minDist = b.radius + b2.radius + 4; if (dist < minDist && dist > 0) { const angle = Math.atan2(dy, dx), overlap = (minDist - dist) / 2; b.x -= Math.cos(angle)*overlap; b.y -= Math.sin(angle)*overlap; b2.x += Math.cos(angle)*overlap; b2.y += Math.sin(angle)*overlap; const tmpVx = b.vx, tmpVy = b.vy; b.vx = b2.vx*0.92; b.vy = b2.vy*0.92; b2.vx = tmpVx*0.92; b2.vy = tmpVy*0.92; } }); ctx.save(); ctx.globalAlpha = b.opacity; ctx.beginPath(); ctx.arc(b.x, b.y, b.radius, 0, Math.PI*2); ctx.fillStyle = b.color; ctx.fill(); ctx.fillStyle = "#ffffff"; ctx.font = `500 ${Math.max(11, Math.floor(b.radius * 0.28))}px 'DM Sans', sans-serif`; ctx.textAlign = "center"; ctx.textBaseline = "middle"; ctx.fillText(b.label, b.x, b.y); ctx.restore(); }); animRef.current = requestAnimationFrame(animate); }; animate(); return () => { cancelAnimationFrame(animRef.current); window.removeEventListener("resize", resize); document.removeEventListener("visibilitychange", handleVisibility); }; }, [initBubbles]); return ( ); } // ─── MAIN COMPONENT ────────────────────────────────────────────────────────── function Portfolio() { const [activeSection, setActiveSection] = useState("Home"); const [activeCategory, setActiveCategory] = useState("ML Apps"); const [scrolled, setScrolled] = useState(false); const [statsVisible, setStatsVisible] = useState(false); const [articles, setArticles] = useState(STATIC_ARTICLES.slice(0,3)); const [mobileMenuOpen, setMobileMenuOpen] = useState(false); const statsRef = useRef(null); function useInView(threshold = 0.12) { const ref = useRef(null); const [inView, setInView] = useState(false); useEffect(() => { const obs = new IntersectionObserver( ([e]) => { if (e.isIntersecting) { setInView(true); obs.disconnect(); } }, { threshold } ); if (ref.current) obs.observe(ref.current); return () => obs.disconnect(); }, []); return [ref, inView]; } const [expRef, expIn] = useInView(0.08); const [projRef, projIn] = useInView(0.06); const [artRef, artIn] = useInView(0.08); const [aboutRef, aboutIn] = useInView(0.08); const [skillRef, skillIn] = useInView(0.1); const [contactRef, contactIn] = useInView(0.1); const [heroRef, heroIn] = useInView(0.01); const [winW, setWinW] = useState(window.innerWidth); useEffect(() => { const fn = () => setWinW(window.innerWidth); window.addEventListener("resize", fn); return () => window.removeEventListener("resize", fn); }, []); const isMobile = winW <= 600; const isTablet = winW <= 900; // ── 3D About Carousel ────────────────────────────────────── const [carouselIdx, setCarouselIdx] = useState(0); useEffect(() => { const t = setInterval(() => setCarouselIdx(p => (p + 1) % 3), 4200); return () => clearInterval(t); }, []); // ── Stacked Article Deck ─────────────────────────────────── const [deckArts, setDeckArts] = useState([0, 1, 2]); const [deckExiting, setDeckExiting] = useState(false); function nextDeckCard() { if (deckExiting) return; setDeckExiting(true); const nextIdx = (deckArts[2] + 1) % STATIC_ARTICLES.length; setTimeout(() => { setDeckArts(prev => [prev[1], prev[2], nextIdx]); setDeckExiting(false); }, 480); } function getDeckStyle(pos) { if (deckExiting) { if (pos === 0) return { ty: 340, scale: 1, zi: 10, op: 0 }; if (pos === 1) return { ty: 0, scale: 1, zi: 3, op: 1 }; return { ty: -20, scale: 0.95, zi: 2, op: 1 }; } if (pos === 0) return { ty: 0, scale: 1, zi: 3, op: 1 }; if (pos === 1) return { ty: -20, scale: 0.95, zi: 2, op: 1 }; return { ty: -44, scale: 0.9, zi: 1, op: 1 }; } const WORK_EXP = [ { role: "Returns Data Admin", company: "Alloga (Cencora / UDG Healthcare)", period: "March 2025 – Present", tags: ["Excel","Power BI","PowerPoint","Data Analysis","Reporting"], bullets: [ "Generate data-driven reports using Excel, Power BI, and PowerPoint to track returns processing metrics, supporting management decision-making with actionable insights.", "Streamline returns administration workflows through data analysis, identifying process inefficiencies and contributing to improved operational accuracy across the supply chain.", ], }, { role: "Data Scientist", company: "Thor Explorations, Nigeria", period: "2021 – 2023", tags: ["Python","Scikit-learn","SQL","Power BI","Anomaly Detection"], bullets: [ "Developed Python and Scikit-learn models for anomaly detection and predictive analysis of laboratory assay data across 3 project sites, reducing manual processing time by 35%.", "Designed and built a centralised database for 3 regional projects using Python and SQL, improving data retrieval efficiency by 58%.", "Built automated Power BI reports with scheduled refreshes, replacing manual reporting and saving approximately 8 hours per week.", ], }, ]; useEffect(() => { const fn = () => { setScrolled(window.scrollY > 60); const sections = ["home","about","experience","projects","articles","skills","contact"]; for (const id of sections) { const el = document.getElementById(id); if (el) { const rect = el.getBoundingClientRect(); if (rect.top <= 100 && rect.bottom >= 100) { setActiveSection(id.charAt(0).toUpperCase() + id.slice(1)); break; } } } }; window.addEventListener("scroll", fn); return () => window.removeEventListener("scroll", fn); }, []); useEffect(() => { const observer = new IntersectionObserver( ([entry]) => { if (entry.isIntersecting) setStatsVisible(true); }, { threshold: 0.2 } ); if (statsRef.current) observer.observe(statsRef.current); return () => observer.disconnect(); }, []); useEffect(() => { fetch("https://api.rss2json.com/v1/api.json?rss_url=https://medium.com/feed/@Fehintiti&api_key=public") .then(r => r.json()) .then(data => { if (data.status === "ok" && data.items?.length) { const mapped = data.items.slice(0,3).map(item => ({ title: item.title, date: new Date(item.pubDate).toLocaleDateString("en-GB", { month:"short", year:"numeric" }), desc: item.description.replace(/<[^>]*>/g,"").slice(0,160)+"...", url: item.link, })); setArticles(mapped); } }) .catch(() => {}); }, []); const scrollTo = (id) => { document.getElementById(id)?.scrollIntoView({ behavior:"smooth" }); setMobileMenuOpen(false); }; const NAV_LINKS = ["Home","About","Experience","Projects","Articles","Skills","Contact"]; const photoSrc = window.PORTFOLIO_PHOTO || (window.TOMISIN_JPG) || "https://ui-avatars.com/api/?name=Tomisin+Okunlola&size=600&background=f0ebe3&color=222222&bold=true&font-size=0.28"; const carouselImages = [ photoSrc, "https://images.unsplash.com/photo-1551288049-bebda4e38f71?w=600&q=80", "https://images.unsplash.com/photo-1460925895917-afdab827c52f?w=600&q=80", ]; const c15 = useCounter("15",1600,statsVisible); const c3 = useCounter("3", 1600,statsVisible); const c6 = useCounter("6", 1600,statsVisible); const SOCIAL = [ { icon:, url:"https://github.com/Fehintiti", label:"GitHub" }, { icon:, url:"https://www.linkedin.com/in/fehintiti/", label:"LinkedIn" }, { icon:, url:"https://medium.com/@Fehintiti", label:"Medium" }, { icon:, url:"https://twitter.com/fehintiti", label:"Twitter" }, { icon:, url:"mailto:tokunlola12@gmail.com", label:"Email" }, ]; return (
{/* ── NAV ─────────────────────────────────────────────── */} {/* ── HERO ─────────────────────────────────────────────── */}
{/* Vertical sidebar label — desktop only */} {!isMobile && (
Data Scientist · 2026
)} {/* Left/bottom content */}
{/* Stats row */}
{[ { n:c15, suffix:"+", label:"Projects built" }, { n:c3, suffix:"+", label:"Years experience" }, { n:c6, suffix:"", label:"Articles published" }, ].map((s,i)=>(
{s.n}{s.suffix}
{s.label}
))}
{/* Big heading + sub */}

Hello,

— I'm Tomisin Okunlola, Data Scientist

ML models · Cloud pipelines · Business analytics

{/* Scroll indicator */} {!isMobile && (
Scroll down ↓
)} {/* Mobile social row */} {isMobile && (
{SOCIAL.map((s,i)=>( {s.icon} ))}
)}
{/* Photo — right on desktop, top on mobile via order:-1 */}
Tomisin Okunlola {!isMobile && (
{SOCIAL.map((s,i)=>( {s.icon} ))}
)}
{/* ── ABOUT ─────────────────────────────────────────────── */}
{/* Left: text */}

About Me

Data Scientist.
Problem Solver.

MSc in Big Data Analytics. 3+ years of experience across geoscience, logistics, and pharmaceutical supply chain.

I build predictive models, design interactive dashboards, and manage end-to-end data projects — proficient in Python, SQL, Power BI, and cloud pipelines.

{[ "End-to-end ML model development and deployment", "Serverless cloud data pipelines at scale", "Business intelligence that drives decisions", ].map((pt,i)=>(

{pt}

))}
{/* Center: stat card + 3D carousel */}
35%

Reduction in manual processing time achieved at Thor Explorations through ML model automation

{/* 3D Image Carousel */}
{carouselImages.map((src, i) => { const total = 3; const isActive = i === carouselIdx; const isLeft = i === (carouselIdx - 1 + total) % total; const style = isActive ? { transform:"translateX(0px) translateY(0px) scale(1) rotateY(0deg)", opacity:1, zIndex:3, boxShadow:"0 12px 40px rgba(0,0,0,0.18)" } : isLeft ? { transform:"translateX(-52px) translateY(-28px) scale(0.8) rotateY(14deg)", opacity:0.75, zIndex:2, boxShadow:"0 6px 20px rgba(0,0,0,0.1)" } : { transform:"translateX(52px) translateY(-28px) scale(0.8) rotateY(-14deg)", opacity:0.75, zIndex:2, boxShadow:"0 6px 20px rgba(0,0,0,0.1)" }; return ( setCarouselIdx(i)} style={{ position:"absolute", inset:0, width:"100%", height:"100%", objectFit:"cover", objectPosition:"top center", borderRadius:16, cursor:"pointer", transition:"all 0.8s cubic-bezier(0.4,2,0.3,1)", ...style }} /> ); })}
{[0,1,2].map(i => (
setCarouselIdx(i)} style={{ width:6, height:6, borderRadius:"50%", background:i===carouselIdx?"#222222":"#DDDDDD", cursor:"pointer", transition:"background 0.3s" }} /> ))}
{/* Right: second stat + bullet points */}
Tomisin
58%

DB efficiency improvement via centralised pipeline

{[ { title:"University of Derby", sub:"MSc Big Data Analytics — Distinction, 2023–2024" }, { title:"University of Lagos", sub:"BSc Geoscience, Geology — 2015–2020" }, { title:"Open to Opportunities", sub:"UK · Remote · Canada relocation" }, ].map((item,i)=>(
{item.title}
{item.sub}
))}
{/* ── EXPERIENCE ──────────────────────────────────────────── */}

Career

Where the work
got done.

From mining operations in Nigeria to pharmaceutical supply chains in the UK — real roles, real data, measurable outcomes.

{WORK_EXP.map((job,i)=>(
{/* Left meta */}
{job.period}
{job.tags.map(t=>( {t} ))}
{/* Right content */}

{job.role}

{job.company}

{job.bullets.map((b,bi)=>(

{b}

))}
))}
{/* ── PROJECTS ──────────────────────────────────────────── */}

Selected Work

Real data.
Measurable results.

{Object.keys(PROJECTS).map(cat=>( ))}
{/* Featured card */} {PROJECTS[activeCategory][0] && (

Featured Project

{PROJECTS[activeCategory][0].title}

{PROJECTS[activeCategory][0].desc}

{PROJECTS[activeCategory][0].stack.slice(0,5).map(t=>( {t} ))}
{PROJECTS[activeCategory][0].live && Live App } {PROJECTS[activeCategory][0].github && GitHub } {PROJECTS[activeCategory][0].linkedin && View Project }
{PROJECTS[activeCategory][0].title}
)} {/* Grid cards */}
{PROJECTS[activeCategory].slice(1,3).map((p,i)=>(
{p.title}

{p.title}

{p.desc}

{p.stack.slice(0,3).map(t=>({t}))}
{p.github && GitHub } {p.live && Live } {p.linkedin && View Project }
))}
{/* See all on GitHub */}

15+ projects across ML, cloud, computer vision and business intelligence.

See all on GitHub
{/* ── ARTICLES ──────────────────────────────────────────── */}

Writing

The work,
documented.

View all on Medium ↗
{/* ── Stacked Article Deck ── */}
{deckArts.map((artIdx, pos) => { const art = STATIC_ARTICLES[artIdx]; const { ty, scale, zi, op } = getDeckStyle(pos); return (
0{artIdx + 1}
{art.date} {artIdx === 0 && FEATURED}

{art.title}

{art.desc}

Read on Medium
); })}
{/* Controls */}
{deckArts[0] + 1} / {STATIC_ARTICLES.length}
{STATIC_ARTICLES.map((_,i) => (
))}

6 published articles — all projects, walkthroughs, and case studies.

See all articles on Medium
{/* ── SKILLS ──────────────────────────────────────────────── */}

Stack

The tools I use
to build and ship.

{/* ── CONTACT ─────────────────────────────────────────────── */}

Contact

Let's build something
that matters.

Open to data science and data engineering roles. Available for collaborations, consulting, and technical conversations.

{[ { label:"Email", val:"tokunlola12@gmail.com", href:"mailto:tokunlola12@gmail.com" }, { label:"LinkedIn", val:"@fehintiti", href:"https://www.linkedin.com/in/fehintiti/" }, { label:"GitHub", val:"@Fehintiti", href:"https://github.com/Fehintiti" }, { label:"Medium", val:"@Fehintiti", href:"https://medium.com/@Fehintiti" }, ].map((item,i)=>(
{item.label}
{item.val}
 ))}

© 2026 Tomisin Okunlola · Built with React

{SOCIAL.map((s,i)=>( {s.icon} ))}
); } const _root = ReactDOM.createRoot(document.getElementById('root')); _root.render(React.createElement(Portfolio));