Hi, I'm Sam Adeyemi
Software Engineer — Generative AI
85%
Faster workflows
73%
Fewer errors
6+
Years experience
Building production AI systems that transform complex ideas into scalable software.
01 / ABOUT
Who I Am
I'm a Software Engineer — Generative AIwith 6+ years of experience building production systems. My work focuses on Python-led APIs, metadata-driven workflows, and validation pipelines that deliver measurable impact.
I've built systems that reduced output errors by 73% and accelerated generation workflows by 85%. My approach combines rigorous software engineering with a practical, commercial mindset.
Currently, I'm focused on Generative AI engineering — turning early-stage AI ideas into secure, scalable applications that solve real-world challenges.
Configuration over Code
JSON/YAML-driven architecture
Validation-First
73% fewer errors
Event-Driven Orchestration
85% faster workflows
Quick Facts
02 / SERVICES
What I Deliver
I build systems that solve real problems — with measurable results.
AI-Powered Workflows
Event-driven orchestration and metadata-driven pipelines that accelerate generation by 85%.
Production APIs
FastAPI, Python, REST APIs with validation layers and modular service boundaries.
Validation Pipelines
Schema enforcement and rule-based validation that reduced output errors by 73%.
Generative AI Integration
AI-assisted engineering with structured prompt/output review and governance.
Metadata-Driven Architecture
JSON/YAML-driven systems where non-technical users can adjust behaviour without code.
Full-Stack Foundations
Vanilla JavaScript, React-transferable components, API integration, and responsive UI.
03 / IMPACT
Measurable Results
These aren't just numbers — they're proof of systems that work.
85%
Faster generation via parallelised orchestration
73%
Fewer errors through schema validation pipelines
20+
Automated validation rules built (TDCT Validator)
6+
Years of production software engineering experience
04 / PROJECTS
Featured Work
Systems I've built — each one solving a real problem with measurable outcomes.
247 orgs • ₦2.4B+ tracked • Live in production
⭐ SeedTracker
Multi-tenant fintech SaaS platform for organisational financial management — dual-gateway payments, offline-first sync, WhatsApp-native notifications and AI engagement predictions.
6 agents • 85% faster • 73% fewer errors
⭐ PHHM Multi-Agent AI Platform
A modular Generative AI platform built around six specialised agents — Overseer, Welcome, Analyst, Care, Communications and Gospel — collaborating through a central orchestration layer.
10+ tools • 100% AI-driven
AI CV Builder Pro™
Full-stack AI career system with 10+ specialised tools — CV Builder, Interview Coach, Company Intelligence, Salary Negotiator, and more.
500+ users • 4.8/5 rating • GPT-4
Alpha Wings AI — Post Booster
GPT-4-powered SaaS that generates and adapts social media content across four platforms from a single input, on a tiered subscription model.
20+ rules • MIT-licensed
TDCT Validator
Python/YAML validation framework running 20+ automated checks on clinical trial protocols. Detects logical, safety, and design flaws before trial initiation.
AI-assisted • Clinical documentation
TrialForge AI
AI-powered tooling supporting clinical trial documentation, validation and protocol automation for research teams.
05 / PUBLICATIONS
Published Research
Peer-reviewed research at the intersection of AI governance, ethics, and validation systems.
Governance and Legitimacy in AI: Developing and Testing the AI Authorship Legitimacy Model
Springer Nature — AI and Ethics, 2026
Automated Trial Design Consistency Testing (TDCT): A Validation System for Detecting Protocol Design Flaws Before Clinical Trial Initiation
Zenodo Preprint, 2025
06 / HIGHLIGHTS
Professional Highlights
A snapshot of academic achievement, published research, and production engineering work.
First Class BEng (Hons) Software Engineering
MSc (Distinction)
Published in Springer Nature — AI & Ethics
Peer Reviewer — Springer Nature (Trials Journal)
Designed a production Multi-Agent AI Platform
Built multiple AI-powered SaaS platforms
Author of the 15-part PHHM Journal on production AI engineering
07 / WRITING
The PHHM Journal
A 15-part engineering series on building production multi-agent AI systems — architecture, orchestration, guardrails, observability, cost engineering, and the war stories in between. Written from real production experience, not theory.
Why We Didn't Use LangChain in Production (And What We Built Instead)
Evaluating LangChain, CrewAI, and AutoGen taught us an important lesson — use frameworks for capabilities, but own th…
The Production Incident That Changed How We Deploy AI Prompts
No alert fired. Nothing crashed. A post-mortem on the single sentence that quietly degraded our Care Agent — and how…
Building a Multi-Agent AI Orchestration Layer in Python
Lessons from six collaborating agents — how PHHM reduced workflow time by 85% and errors by 73% with specialist agent…
15 in-depth articles covering the full production AI stack — best read in order.
08 / CONTACT
Let's Build Something Great
Available immediately. Open to remote and UK-based roles.
© 2026 Sam Adeyemi. Built with Next.js, TypeScript & Tailwind CSS.