Full‑stack & AI/ML engineering that ships.
I’m Abiodun Caleb Adesesan. I build production-ready web products and practical ML prototypes—focused on clarity, performance, and measurable outcomes.
Who I am
I am Abiodun Caleb Adesesan. I'm a full‑stack engineer with an AI/ML focus. I help teams ship clean, fast web products and practical ML prototypes—from UI to APIs to deployment.
Strengths: crisp UI hierarchy, performance-aware implementation, and communication that keeps projects moving.
Case studies (high-signal)
A few projects shown the way recruiters and clients read them: problem → approach → outcomes.
View all on GitHub →Web Design & Build · 2026
Paris American International University
A university site needed a modern, mobile-first presence with clear program discovery and a maintainable admin workflow.
OpenWordPressElementorPerformance optimizationSEO basicsClick for details
Web Design & Build · 2026
Paris American International University
A university site needed a modern, mobile-first presence with clear program discovery and a maintainable admin workflow.

Approach
- Designed an information architecture focused on prospective-student questions.
- Implemented responsive templates in WordPress + Elementor with reusable sections.
- Optimized assets and layout for fast first load and stable CLS.
Challenges
- Balancing marketing-style visuals with academic clarity.
- Keeping templates flexible for non-technical updates.
Outcomes
- Clearer navigation and faster page interactions on mobile.
- Easier content updates for admins via modular sections.
Conversion-focused Website · 2026
Study in North Cyprus
An education consultancy needed a site that builds trust quickly and guides visitors to inquiry actions.
OpenWordPressElementorUI/UXAnalytics-ready structureClick for details
Conversion-focused Website · 2026
Study in North Cyprus
An education consultancy needed a site that builds trust quickly and guides visitors to inquiry actions.

Approach
- Built a homepage that leads with value proposition and clear CTAs.
- Created landing-page style sections for services, proof, and FAQs.
- Improved readability and hierarchy across devices.
Challenges
- Ensuring the content stays scannable without losing detail.
Outcomes
- Stronger CTA visibility and a clearer inquiry path.
Applied ML (NLP) · 2025
PEFT / LoRA — Sentiment Analysis
Fine-tune a transformer for sentiment classification efficiently (lower compute/cost) while tracking baseline vs tuned performance.
OpenPythonHugging Face TransformersPEFT/LoRAJupyterClick for details
Applied ML (NLP) · 2025
PEFT / LoRA — Sentiment Analysis
Fine-tune a transformer for sentiment classification efficiently (lower compute/cost) while tracking baseline vs tuned performance.
Approach
- Set up a baseline DistilBERT evaluation on IMDB.
- Applied LoRA via PEFT and compared metrics against the baseline.
- Documented the experiment steps for reproducibility.
Challenges
- Keeping training stable while remaining compute-efficient.
Outcomes
- A repeatable workflow for efficient fine-tuning experiments.
Product / Commerce UI · 2025
Gourmet Bakes
Create a clean, appetizing commerce experience with strong hierarchy, mobile-first layout, and conversion-friendly CTAs.
OpenTypeScriptReact UI patternsResponsive designClick for details
Product / Commerce UI · 2025
Gourmet Bakes
Create a clean, appetizing commerce experience with strong hierarchy, mobile-first layout, and conversion-friendly CTAs.

Approach
- Designed a homepage layout optimized for scanning categories and key actions.
- Built reusable UI sections and components for rapid iteration.
- Focused on contrast, spacing, and responsive typography for readability.
Challenges
- Balancing visual richness with speed and layout stability.
Outcomes
- A modern UI foundation suitable for a real productized food/commerce build.
Additional repos and builds. I keep this list curated—quality over quantity.
Professional Client Work · WordPress · ElementorParis American International University
Developed a modern, visually engaging web presence for the Paris American International University. Built using WordPress and Elementor to ensure a highly responsive, optimized, and easily manageable platform for the institution.
Professional Client Work · WordPress · ElementorStudy in North Cyprus
Educational consultancy platform crafted to guide international students. Custom-built with WordPress and Elementor for a seamless, conversion-focused user experience with modern aesthetics.
- Python · Jupyter · Hugging Face
PEFT / LoRA — Sentiment Analysis
Parameter-efficient fine-tuning (LoRA) on DistilBERT for IMDB sentiment classification. Baseline vs fine-tuned evaluation with Transformers & PEFT.
TypeScriptGourmet Bakes
Modern TypeScript codebase for a bakery / food commerce experience — structure and UI craft tuned for real-world use.
- JavaScript
FinLogger Dashboard
Finance logging and dashboard UI — data-forward interface patterns and client-side interactivity.
PHPE-commerce Website
Full e-commerce site — catalog, flows, and server-side patterns in PHP for production-style delivery.
JavaScript · ArchitectureStudy Planner — Microservices
Study planner built around microservices thinking — decomposition, services, and integration across the stack.
Professional Project · JavaScriptContact Management
Contact management system developed at a professional architecture level — featuring CRUD operations, record organization, and practical, production-ready UI.
How I can help
Choose the lane you need—product build, UI polish, or an AI/ML prototype—then we scope it into a clean, shippable plan.
Full‑stack product builds
From idea to shipped feature — clean UI, reliable APIs, and production-ready delivery.
- Next.js / React apps
- API design (REST)
- Performance + accessibility
AI/ML prototyping
Experiment-driven ML work — evaluation, iteration, and pragmatic integration into products.
- NLP workflows
- Fine‑tuning (PEFT/LoRA)
- Data-first iteration
UI/UX engineering
Design-to-code execution that feels polished: motion, hierarchy, responsiveness, and clarity.
- Component systems
- Interaction design
- Mobile-first layout
Process is everything
A simple, streamlined workflow that keeps scope clear and momentum high.
Step 1
Align on goals
We define the outcome, audience, constraints, and success metrics — then pick the fastest path to impact.
Step 2
Design the solution
Information architecture, UI structure, and technical approach — optimized for clarity and maintainability.
Step 3
Build & iterate
Ship in slices, validate quickly, and keep quality high (performance, accessibility, tests where needed).
Step 4
Launch & improve
Handoff docs, analytics readiness, and follow-up improvements based on real usage.
Recent highlights
A quick gallery of recent work—design, engineering, and outcomes in one glance.
Why partner with me
What working together feels like: clear communication, strong UI taste, and delivery that stays fast and predictable.
01
Clarity first
I make it obvious what you do, who it’s for, and what to do next—especially on mobile.
02
Design + engineering
Polished UI and motion, but implemented with performance, accessibility, and maintainability in mind.
03
Fast iteration
I ship in slices, get feedback early, and avoid big-bang surprises late in the project.
04
Pragmatic AI/ML
Experiment-driven ML work that integrates cleanly into products—evaluation and trade-offs included.
05
Reliability
Clean handoff, documented decisions, and an approach that keeps momentum high.
06
Business outcomes
Every section exists to support a goal: trust, clarity, conversion, or decision-making.
Quick stats
Shipped sites
10+
Client + personal builds
Featured case studies
4
Problem → approach → outcomes
Response time
24–48h
Typical reply window
Clients love the work
A few short notes that reflect how I collaborate: clear, reliable, and detail-oriented.
“Clear communication, fast turnaround, and the final site felt premium on both desktop and mobile.”
Client partner
Education
“Strong sense of UI detail, but also pragmatic about performance and maintainable implementation.”
Teammate
Engineering
Experience & achievements
A quick timeline of what I’ve been building. Full details live on LinkedIn.
Software engineering & product builds
PresentIndependent / academic
Shipping full-stack web apps, dashboards, and ML experiments — from TypeScript frontends to PHP backends and Jupyter research workflows.
Applied NLP — efficient fine-tuning
Research & MLPEFT / LoRA project
DistilBERT sentiment analysis with LoRA; evaluation against baselines using Hugging Face Transformers.
Distributed systems mindset
ArchitectureMicroservices study planner
Designing planner workflows across services — practical introduction to service boundaries and integration.
Stack & craft
Grouped by what I use to ship: languages, web/UI, ML/data, and engineering practices. More history on LinkedIn.
Languages
- TypeScript
- JavaScript
- Python
- PHP
Web & UI
- React
- Next.js
- Tailwind CSS
- HTML5 Canvas
- Framer Motion
- WordPress
- Elementor
ML & data
- Hugging Face Transformers
- PEFT / LoRA
- Jupyter
- NLP workflows
Practices
- Microservices
- REST APIs
- Git / GitHub
- Performance-aware UI
Notes & experiments
Short posts that show how I think—trade-offs, experiments, and lessons learned.
2026
Efficient fine‑tuning with LoRA: what to track
A practical checklist for evaluating PEFT runs: baseline setup, metrics, and common failure modes.
2026
Designing dashboards that scan well
Hierarchy, whitespace, and interaction patterns that make data UIs feel effortless to use.
Questions, Answers
Get quick answers to your most pressing questions.
Client partner
Education
“Clear communication, fast turnaround, and the final site felt premium on both desktop and mobile.”
Software engineering roles (frontend/full‑stack) and freelance builds where you need polished UI, reliable APIs, and clear delivery.
Let's build incredible products together.
Two clear starting points—pick what fits. We can refine scope on a quick call.
Full‑stack web build
Starting from $899High-performance Next.js/React builds tailored for clarity, speed, and SEO-ready structure.
AI/ML prototype
Starting from $1,299Applied NLP or lightweight ML features with evaluation, iteration, and integration guidance.

Featured preview
A quick visual reference for what you’ll get—polish, hierarchy, and conversion-ready UI.
Let’s build something together
Send a message or book a call. I usually respond within 24–48 hours.
Prefer direct email?
abiodunadesesan@gmail.com
Share a short overview, timeline, and links (if any). I’ll reply with next steps and a simple plan.
Open email draft →Availability
Open to engineering roles, freelance projects, and collaborations.