The work · By company, documented and ranked

15+ projects, delivered end to end.

The deep dives below were written against the actual code: module inventories, real routes, git history. Where something was a mockup, synthetic, or unvalidated, the page says so. Client names, countries, and figures are out. My engineering results are in.

Judge validated at Cohen's k=0.881Faithfulness 0.984, leakage measured at zero
Voice agents in five languagesFrom SIP trunking to local model serving
Government, telecom, enterpriseDelivered under private-network and data-residency constraints

Client work

Strategy&
· PwC

strategyand.pwc.com

Government and enterprise engagements, anonymized under the rules on each page. Ranked deliberately: a portfolio that presents everything as equally impressive is less credible than one that ranks its own work.

Government / Public Health (GCC)Grounded Strategy Intelligence Workspace

Highest in portfolio - ~17k LOC, 6-stage ingestion, evidence-gated scoring, validated LLM-judge evaluation

Flagship ragvision-llmevaluationjudge-calibration
Read the deep dive
Telecommunications (GCC) - consumer business unitCommercial Command Tower

High - closed-loop analytics, statistical anomaly detection, report generation, RBAC

Flagship analyticsanomaly-detectionrbacllm-explanations
Read the deep dive
Professional services - internal intelligence product (MENA/GCC coverage)Media Intelligence Platform

High - multi-stage NLP pipeline, 4-tier LLM funnel, adaptive vector clustering, MinHash dedup, bilingual entity resolution

Flagship nlp-pipelinevector-searchclusteringentity-resolution
Read the deep dive
Sovereign wealth / public investment (GCC)Sovereign Investment Country Dashboard

Medium-high - three-tier executive information architecture, ~20 bespoke chart components, dense comparative visualisation

Substantial delivery executive-dashboardinformation-architecturedata-visualisationcustom-charting
Read the deep dive
Real-estate investment / pension-fund-linked portfolio (GCC)Portfolio Performance Monitoring Agent

Medium-high - schema inference from unstructured workbooks, real tool-calling agent loop, privacy-constrained handling

Substantial delivery agentexcel-intelligenceschema-inferencedashboard
Read the deep dive
Government - agriculture, environment and water (GCC)National Strategy Delivery Dashboard

Medium - bilingual RTL data model, many-to-many objective/initiative mapping, generated data layer

Focused contribution governmentstrategy-monitoringbilingualrtl
Read the deep dive
Government - economic development / free zones (GCC)Geospatial Economic Zones Intelligence Demo

Medium-high - offline-first GIS, packaged vector basemap, real geospatial analysis, test-enforced network invariant

Substantial delivery gisgeospatialoffline-firstvector-tiles
Read the deep dive
Professional services - public-facing capability assistantEnterprise Capability Knowledge Chatbot

Medium - vendor-delivered platform; my contribution was quality assurance and safety review

Focused contribution ragprompt-engineeringquality-assuranceprompt-injection-defence
Read the deep dive
Real-estate investment trusts / financial analysis (GCC)Financial Document Extraction & Benchmarking Tool

Medium-high - OCR and native PDF handling, table extraction, field mapping, accuracy benchmarking

Substantial delivery document-extractionocrstructured-outputbenchmarking
Read the deep dive
Personal engineering infrastructure (self-directed R&D)Agentic Executive Assistant & Portfolio Command System

High - multi-agent orchestration, MCP tool integration, always-on daemon, native macOS app

Flagship agentic-systemsmcpmulti-agentdaemon
Read the deep dive
Government (multiple entities) and public-sector PMO - GCC/MENAStructured Data Query Agent Family

Medium - repeated and hardened across four engagements

Substantial delivery agentstool-callingnatural-language-querypmo
Read the deep dive

Independent builds

My own
products

Built on my time, run on my money.

Personal productViral Alchemy

A system that maps what is about to trend: niche sources scanned, engagement clustered, ideas turned into scripts.

viralalchemy.io

The Learning EngineerWebinar campaign, run by AI

A full webinar launch for my education brand: AI wrote the scripts for three video ads, built the slides, the funnel, and the landing page, and ran Meta campaigns through a connector I built before an official one existed, monitoring every 12 to 24 hours. Cost per lead: 17 cents. 2,000+ registrations. Around 1,700 replay views on YouTube. Two to three weeks, next to a full-time job, first attempt.

Watch the workshop · @thelearning.engineer

Freelance + prospect workChatbots and avatars

A Telegram assessment chatbot with RAG and webhook shipment tracking (UK client). A real-time avatar demo built in days, deployed near the APIs for latency; the prospect called it magic and hired someone else. Both true.

Earlier · 2022 to 2025

Artefact
· Innovum

artefact.com

The data-science years: NLP topic classifiers and sentiment models for multinational clients, geographic clustering for charger infrastructure, transcription and summarization pipelines. Where the discipline of measuring before claiming was learned.

The first step

Twenty minutes. You describe where AI is stuck.

You leave the call knowing whether I can help, roughly what it would take, and what it would cost to find out for sure. If I am not the right person, I will say so on the call.

The 20 minutes, in order

  1. You talk first. Where AI is stuck, what has been tried, what it costs today.
  2. I answer plainly. Whether I can help, and what I would look at first.
  3. You leave with a next step. An audit scope, a pointer elsewhere, or a clean no.

Before you book

  1. Your stack is not too messy to start. Messy is the normal starting condition.
  2. Training that does not survive the week is the normal outcome. These sessions build on your backlog and ship something real.
  3. You do not need budget approved to take the call. You need it approved to start step two.