Our work: AI products and data platforms
Three systems in production, and data, cloud and AI case studies.
AI product · In production
Answerable
Our AI working environment for proprietary knowledge and data, in production and daily use. Its first deployment turned a research business’s archive and forecast data into analysis people can trust and trace back to source.
78,500+ publications89,000 chart assets35,700 forecast points
Read the case study: Answerable See Answerable Answerable
+ New chat
Today
What are the key growth…
What regulatory or p…
ChatSearchSummariesCompare
Grounding check passed · 3 sources · house view current
Sources
[1]AI boom analysis, June 2026Compute and data-centre spend as an earnings driver
[7]Semiconductor sector noteCPU and high-bandwidth memory demand
[6]China quarterly outlookTelecoms investment and data-centre build
- The problem
- Decades of proprietary research, forecast data and chart assets, and customers who needed complex questions answered from the organisation’s current view, with the evidence open to inspection.
- What we built
- Retrieval, synthesis and structured forecast data combined into source-grounded analysis. Recency rules keep answers on the current view, a separate check tests each answer against its sources, and every point carries a reference a reader can open. TechGuidr led the product design, architecture, data, engineering and interface.
- The proof
- In production and daily use across more than 78,500 publications. Along the way we found premium research publicly reachable through a content-delivery misconfiguration and reported the fix to leadership.
- The evidence
- Measured. The size of the archive in the first deployment. No usage or value figures are published.
AI product · Commercial intelligence
Foreseeable
Know which accounts need you, and why. Commercial signals turned into next actions, each with its reasoning, evidence and confidence.
Read the case study: Foreseeable
- The problem
- Account signals, renewal risk and portfolio information sat in different systems, so warnings arrived late and account planning relied on individual interpretation.
- What we built
- Portfolio intelligence, renewal risk, diagnostics and personalised actions in the team’s daily workflow. Calculations stay deterministic; AI does the inference and drafting, never the arithmetic. Every action carries its reasoning and evidence, a person approves or rejects it, and the AI can be paused with one click.
- The proof
- In production and used daily by the sales and marketing teams.
- The evidence
- Qualitative. Daily use by the sales and marketing teams. No measured sales uplift is claimed.
Platform rebuild · Learning and wellbeing
SaaS Rebuild
4–5 s → <600 msDashboard queries
Read the case study: SaaS Rebuild
- The problem
- Ambitious product plans on a fragmented estate, with dashboards disconnected from live data and calculations that did not follow the agreed methodology.
- The approach
- Bring the fragmented applications, analytics and cloud infrastructure onto one platform, with AI agents built in.
- What was delivered
- Both applications and the iOS and Android apps rebuilt on a Databricks lakehouse, with verified calculation pipelines and an AI coach, identity and deployment controls strengthened, and the business migrated tenant to tenant.
- The outcome
- Dashboard queries went from four to five seconds to under 600 milliseconds once an unnecessary middleware layer was removed.
- The evidence
- Measured. Dashboard query times before and after the rebuild.
Case studies
Data and cloud platform case studies
Data, cloud and analytics work for an aerospace manufacturer, a university, games studios, a retailer, a national charity and a betting company. Each sets out the problem, the approach, what was delivered and the outcome.
Aerospace · Data platform
Real-time intelligence for a global engine fleet
A global aerospace manufacturer
- The problem
- One of the world's largest aerospace companies needed to see its commercial airline engine fleet as it flew, not in retrospective reports.
- The approach
- Analyse petabytes of engine data in the cloud for patterns and anomalies.
- What was delivered
- A cloud data platform on Microsoft Azure and Power BI.
- The outcome
- A real-time view of the fleet, with the ability to predict engine maintenance events early enough to minimise aircraft time out of service.
Higher education · Cloud
A university sent home overnight
A leading UK university
- The problem
- When the pandemic hit, a leading UK university needed more than 5,000 staff and students working and learning remotely, at speed, on infrastructure never designed for it.
- The approach
- A modern desktop platform that could scale quickly, rolled out under tight time pressure.
- What was delivered
- Azure Virtual Desktop for more than 5,000 staff and students.
- The outcome
- Remote teaching and working at full scale, on a platform designed to outlast the crisis that forced it.
Gaming · Analytics
Studio analytics in the early days of Azure
AAA games studios
- The problem
- AAA games studios each held their own analytics, with no central view across titles and platforms, at a time when Azure itself was young.
- The approach
- Build on a young Azure platform, dogfooding the internal Microsoft tooling that became public services including Power BI.
- What was delivered
- Studio analytics centralised on Azure, and Global BI workloads migrated from on-premises to Azure.
- The outcome
- A single analytics estate across studios, built early enough that the tools it proved out went on to become products.
Retail · Data warehouse
Finding where the stock was going
A leading UK retailer
- The problem
- Significant, unexplained losses in stock management, with the evidence buried in high-volume transactional data.
- The approach
- Record the patterns and trends in stock movement, with automated analysis around the events where losses clustered.
- What was delivered
- A data warehouse platform on SQL Server for the high-volume transactional data.
- The outcome
- The stock and logistics lifecycle segmented and visible, so losses could be traced to causes rather than absorbed as a cost of trading.
Non-profit · Cloud platform
Tracking every donated bag
A national charity
- The problem
- Donated goods were leaking out of the collection network before reaching stores, and the retail side of the charity ran with no regular, accurate analysis.
- The approach
- Bring collections, stores and sales into one view: legacy tills connected for real-time sales, and IoT tracking across vehicles, textile banks and clothing bags.
- What was delivered
- A greenfield Azure data platform with a 360-degree view of operations.
- The outcome
- A consolidated view of donated goods, collection operations and retail sales.
Gambling & gaming · Real-time data
Watching the odds being played in real time
A betting company
- The problem
- Fixed-odds terminals and spread betting carried fraud patterns the company could only see after the money had gone.
- The approach
- Reveal, in real time, who was playing the odds on a particular event.
- What was delivered
- A cloud data platform, and a Data & Advanced Analytics practice with the architects, engineers and data scientists recruited around it.
- The outcome
- Fraudulent activity visible as it happened rather than in the post-mortem, and a data practice that outlived the engagement.