Hi, I'm Harrison!
I automate data & AI workflows
that save teams time. Open to work

I'm a Data and AI Solutions Architect and the founder of Telaq Solutions (opens Telaq Solutions website in a new tab), a data and AI automation agency. By day I'm a Lead data analyst at Brisdoc Healthcare (opens Brisdoc Healthcare website in a new tab). I hold an MSc in Health Data Science from Swansea University. I'm also building Klayr (opens the Klayr website in a new tab), my own MVP.

Book a call

Feel free to explore my portfolio and reach out. I'd love to connect!

Experience & education
  1. Founder & Data and AI Solutions Architect

    Telaq Solutions

    2026 – Present · AI automation agency

  2. AI for Business Transformation

    Primed Talent

    2026 · Certification

  3. Lead Data Analyst

    Brisdoc Healthcare

    2023 – Present

  4. MSc Health Data Science

    Swansea University

    2023 · Postgraduate degree

Stack I use
  • Python
  • SQL
  • Node.js
  • REST APIs
  • OpenAI
  • n8n
  • Make
  • Zapier
  • Supabase
  • Google Workspace
  • Microsoft 365
  • Healthcare analytics
Results & credentials
15m+
saved per tenant application
25h
expected weekly time reclaimed
MSc
Health Data Science
Location

United Kingdom

Working with clients worldwide

How I work

Start with the problem → Involve people → Design the solution

01 Identify the real problem

I identify the genuine business challenges that matter and align them with strategic objectives and measurable outcomes, so we focus on real problems, not just visible ones.

Selected work

A product I built end to end, plus AI and data systems I've designed and built. Open any project to see the problem, my role and the outcomes.

MVPFeatured

Klayr (my MVP)

Solo-built MVP, in testing

The problem
People save how-to videos from YouTube, TikTok and Instagram and never act on them. The ideas pile up without a plan.
My role
Founder and product owner. Defined the product, designed the user experience and built it with AI-assisted development.
Outcomes
  • Shipped a working MVP: mobile app, web app and marketing site
  • Free and Pro plans with metered AI credits
  • Tester pilot of 12 people (6 learners, 6 business owners) informing who to launch to
  • AI product
  • React Native
  • Product strategy
Data & AI

AI Data Analyst: Chat With Your Data

Self-serve insights, no analyst queue

A login platform where non-technical staff ask questions about company data in plain English and get answers and charts.

The problem
Non-technical teams rely on data analysts for every question, which creates bottlenecks. General chat tools aren't connected to company data, and nobody has checked that the data they see is clean or safe to expose.
What I built
Staff log in to a custom chat platform and ask questions in everyday language. An AI agent inspects the schema first, queries the data, checks its maths with a calculator tool and returns the answer with a chart.
My role
Solo builder and data lead. Designed the data schema and cleaning process, built the AI agent workflow in n8n and the front-end platform, drawing on my background as a data analyst.
Outcomes
  • Non-technical staff query data in plain English instead of waiting for an analyst
  • Answers are verified with a calculator tool and shown with auto-generated charts (via QuickChart)
  • Connects to live data (Google Sheets in the demo, built to work with Postgres or MySQL) or an uploaded CSV
  • Governed by design: data is cleaned and documented first, only aggregated, non-identifiable datasets are exposed
  • Natural language processing
  • AI agents
  • n8n
  • Google Sheets
  • SQL
Healthcare

Predictive Patient Care & Outreach Suite

22% expected drop in readmissions

Predictive analytics that flags high-risk patients early and triggers automated outreach.

The problem
Patients can deteriorate unnoticed between visits. Reactive care raises readmission rates and strains clinical resources.
What I built
A system that monitors patient markers in real time, uses early-warning models to flag risk and triggers outreach so teams can intervene sooner.
My role
Data and AI lead at Telaq. Scoped the clinical problem, designed the early-warning approach and automation flow.
Outcomes
  • 22% expected reduction in readmissions
  • Shifts care from reactive to proactive
  • Automated outreach replaces manual follow-up lists
  • Predictive analytics
  • Healthcare
  • Automation
Real Estate

AI-Powered Tenant Screening System

15m+ saved per application

An AI engine that extracts and verifies rental application data into a decision-ready summary.

The problem
Manual background and income checks took 15-20 minutes per application, causing backlogs at peak times, higher vacancy costs and inconsistent screening.
What I built
An AI engine that extracts and verifies application data, calculates debt-to-income ratios and checks property-specific criteria.
My role
Founder and lead builder at Telaq. Designed the screening logic and built the extraction and verification workflow.
Outcomes
  • 15m+ saved per application (from 15-20 minutes of manual checking)
  • Standardised, decision-ready summary for every applicant
  • Managers focus on final approvals, not data entry
  • AI
  • Document extraction
  • Automation
Reputation & Fintech

AI Reputation & Content Engagement System

25h expected weekly time reclaimed

One engine that monitors reviews, analyses sentiment, drafts replies and plans social content.

The problem
Local service businesses lose 10-15 hours a week managing feedback across platforms. Slow, inconsistent replies hurt trust and revenue.
What I built
A unified engine that monitors reviews, analyses sentiment trends, drafts personalised replies and turns praise into a scheduled content calendar.
My role
Founder and lead builder at Telaq. Designed the human-in-the-loop workflow and built the review and content automation.
Outcomes
  • 25h expected weekly time reclaimed
  • Consistent, faster responses across review platforms
  • Humans approve every reply, so AI drafts but people stay in control
  • Sentiment analysis
  • LLM
  • Human-in-the-loop

Automation pipelines

How the systems I've built move data from input to decision. Hover or tap a step.

Chat question1Inspect schema2Query data3Check maths4Answer + chart5

Data in action

Illustrative example (sample data): work shifting from manual to automated over time. Toggle series, hover points.

0125250375500JanFebMarAprMayJunJulAug

Ask my AI

A live demo of the kind of assistant I build.

Hi! I'm an AI assistant trained on this portfolio. Ask me about projects, skills or how to hire Harrison.

Let's automate something

Tell me about the process that eats your team's time.

hpottxx@gmail.com