Skip to content

Everything between an idea and software people rely on.

Software development

Web platforms, back-end systems and internal tools built to run for years.

We take products from first commit to production and stay for the releases after. That means clear architecture, tests that run on every change, and documentation your own team can pick up.

What we do

  • Web applications and customer portals
  • APIs and back-end systems
  • Legacy modernisation and re-platforming
  • Data pipelines and third-party integrations
  • Event-driven systems and message queues
  • High-performance services in Rust and Go
  • Analytics platforms on ClickHouse, with Matomo for privacy-first web analytics

Typical stack

  • TypeScript
  • React
  • Next.js
  • Node.js
  • Go
  • Rust
  • Python
  • PostgreSQL
  • Kafka
  • RabbitMQ
  • Redis
  • ClickHouse

AI and automation

AI-native products, assistants and agents, and automation wired into the systems you already run.

We build AI that does a job in production, not a demo that impresses once. That means picking the right model for the task, grounding it in your own data, measuring its answers before and after launch, and keeping cost, latency and privacy under control.

What we do

  • AI-native applications and copilots
  • Chatbots and voice assistants for support, sales and internal teams
  • AI agents that use your tools and APIs
  • Search and answers over your documents and data (RAG)
  • AI added to existing products, CRMs and back-office systems
  • Workflow and business process automation
  • Document processing: extraction, classification, summarisation
  • Evaluation, guardrails, monitoring and cost control
  • Private, self-hosted and on-premise model deployment

Typical stack

  • Claude
  • OpenAI
  • Gemini
  • Llama
  • Python
  • Vercel AI SDK
  • pgvector
  • Qdrant
  • n8n
  • Temporal

Things we build with AI

Support assistant

Answers customer questions from your help centre and order data, and hands off to a person with the full context when it should.

Sales and CRM copilot

Drafts follow-ups, summarises calls, scores leads and keeps the CRM up to date without manual typing.

Document automation

Reads invoices, contracts and forms, pulls out the fields that matter and routes them to the right system.

Internal knowledge search

One place to ask questions across wikis, drives and tickets, with sources cited in every answer.

Content and media tools

Captioning, tagging, translation and highlight clipping for video libraries and streaming catalogues.

Operations agents

Agents that watch queues, dashboards and inboxes, take routine actions and escalate what they cannot handle.

How we keep AI production-grade

Measured before launch
Every assistant is tested against a set of real questions with expected answers, and re-tested on every change.
Grounded in your data
Answers come from your documents and systems, with sources, so they can be checked.
Guardrails
Limits on what the system may say and do, human approval for sensitive actions, and logs of every decision.
Privacy by design
Your data is not used to train public models. Self-hosted and on-premise models are available where data cannot leave.
Cost and speed under control
The right model for each step, caching and monitoring, so the monthly bill and response times are predictable.

Mobile apps

Native iOS and Android, or cross-platform when it genuinely fits.

We build apps that feel at home on each platform and ship through the stores without drama. You get the release pipeline, crash reporting and analytics along with the app.

What we do

  • iOS with Swift and SwiftUI
  • Android with Kotlin and Jetpack Compose
  • Cross-platform with React Native or Flutter
  • App Store and Google Play release management

Typical stack

  • Swift
  • SwiftUI
  • Kotlin
  • Compose
  • React Native
  • Flutter

TV and connected devices

Apps for the living room on tvOS, Android TV, Fire TV, Roku, Samsung and LG.

Ten-foot interfaces have their own rules: remote-first navigation, strict performance budgets and a different certification process for every store. We have been through them.

What we do

  • Apple TV (tvOS)
  • Android TV and Google TV
  • Amazon Fire TV
  • Roku
  • Samsung Tizen and LG webOS
  • Focus navigation and remote-first UX
  • Store certification and pre-certification audits

Typical stack

  • tvOS
  • Android TV
  • Fire TV
  • Roku BrightScript
  • Tizen
  • webOS

Platform coverage

PlatformStackBuildReal-device QAStore certification
Apple TV (tvOS)Swift, SwiftUI, AVKitYesYesYes
Android TV and Google TVKotlin, Compose for TV, Media3YesYesYes
Amazon Fire TVKotlin on Fire OS, Media3YesYesYes
RokuBrightScript, SceneGraphYesYesYes
Samsung TizenTypeScript, Tizen Web, Lightning.jsYesYesYes
LG webOSTypeScript, webOS TV SDK, Lightning.jsYesYesYes

Other devices, including Vizio, Hisense VIDAA, Titan OS, HbbTV and operator set-top boxes, on request.

Video streaming

Live and on-demand video pipelines, from ingest to player.

Encoding ladders, packaging, DRM, delivery and playback are one system, and the weakest part sets the viewer's experience. We design and run the whole chain, with the metrics to prove it is working.

What we do

  • Live and on-demand pipelines
  • Transcoding and packaging (HLS, DASH, CMAF)
  • DRM and content protection
  • Low-latency delivery and multi-CDN
  • Custom players for web, mobile and TV
  • Quality-of-experience analytics

Typical stack

  • FFmpeg
  • AWS Media Services
  • Mux
  • Shaka Player
  • hls.js
  • ExoPlayer
  • AVPlayer
  • Rust
  • Kafka
  • ClickHouse

Where we work in the chain

  1. 01

    Ingest

    Live contribution or file upload

  2. 02

    Encode

    Adaptive bitrate ladder

  3. 03

    Package

    HLS, DASH, CMAF

  4. 04

    Protect

    DRM and tokenised URLs

  5. 05

    Deliver

    CDN and multi-CDN

  6. 06

    Play

    Web, mobile and TV players

  7. 07

    Measure

    QoE and business analytics

We design, build and run every stage, or take over the ones that are hurting.

Streaming stack

Protocols
HLS, LL-HLS, MPEG-DASH, CMAF, WebRTC for interactive
DRM
Widevine, FairPlay, PlayReady
Encoding and packaging
FFmpeg, AWS MediaLive, AWS MediaConvert, Mux, Shaka Packager
Players
Shaka Player, hls.js, ExoPlayer and Media3, AVPlayer, Bitmovin, THEOplayer
Delivery
CloudFront, Akamai, Fastly, Multi-CDN switching
Monetisation
SVOD, AVOD, FAST, PPV, SSAI and CSAI
Analytics
Mux Data, Conviva, Custom QoE dashboards

Game development

Mobile, web and multiplayer games with the back-ends to run them.

From prototype to live operations: gameplay, tooling, multiplayer services, economies and the analytics that tell you what players actually do.

What we do

  • Unity and Unreal Engine
  • Mobile and web games
  • Multiplayer and matchmaking back-ends
  • Live-ops, economy and analytics
  • Porting and performance optimisation

Typical stack

  • Unity
  • Unreal
  • C#
  • C++
  • WebGL
  • Nakama

DevOps and cloud

Infrastructure, pipelines and observability so releases become routine.

We set up the platform your engineers deploy to and keep it healthy: reproducible infrastructure, fast pipelines, sensible alerts and a cloud bill someone is actually watching.

What we do

  • Cloud architecture on AWS, GCP and Azure
  • Kubernetes and containers
  • CI/CD pipelines
  • Infrastructure as code
  • Observability and SRE practices
  • Cost optimisation

Typical stack

  • AWS
  • GCP
  • Azure
  • Kubernetes
  • Terraform
  • GitHub Actions
  • Prometheus
  • Grafana

Software quality assurance

Test strategy, automation and release QA across web, mobile and TV.

Quality is a process, not a phase at the end. We build the test strategy, automate what should be automated, and run release QA on real devices so problems are found before your users find them.

What we do

  • Test strategy and planning
  • Automation for web, mobile and APIs
  • Performance and load testing
  • Security testing
  • Real-device lab for mobile and TV
  • Pre-certification audits for TV app stores
  • Release QA and regression suites

Typical stack

  • Playwright
  • Appium
  • k6
  • Detox
  • OWASP ZAP
  • BrowserStack

CRM and business systems

Our own CRM, set up around your sales and support process, or integrated with the tools you already run.

We build and operate TenBit CRM, so implementing it is not a hand-off to a reseller. We map your pipeline, migrate your data, connect email, telephony, billing and marketing tools, and train the team that will live in it every day.

What we do

  • CRM implementation and configuration
  • Data migration from spreadsheets or legacy CRMs
  • Integrations with email, billing, telephony and marketing tools
  • Custom modules, workflows and reports
  • Hosting in our cloud or on your infrastructure
  • Training and ongoing support

Typical stack

  • TenBit CRM
  • REST API
  • Webhooks
  • PostgreSQL
  • RabbitMQ
  • SSO
  • AI assistants

Product design

Research, UX and interface design that engineers can build as drawn.

Design and engineering sit in the same room here. Flows are tested with real users, systems are specified down to the token, and nothing is handed over that cannot be built.

What we do

  • User research and discovery
  • UX flows and prototyping
  • Interface design and design systems
  • Accessibility reviews

Typical stack

  • Figma
  • Design tokens
  • Storybook
  • Matomo

Technology we work with

Chosen for the job, not for the CV. We pick boring, proven tools by default and reach for Rust, Kafka or ClickHouse when the load or the latency calls for it.

Languages
  • TypeScript
  • JavaScript
  • Go
  • Rust
  • Python
  • Swift
  • Kotlin
  • C#
  • C++
  • BrightScript
Web front-end
  • React
  • Next.js
  • Vue
  • Tailwind CSS
  • Storybook
Mobile and TV
  • SwiftUI
  • Jetpack Compose
  • React Native
  • Flutter
  • tvOS
  • Android TV
  • Roku SceneGraph
  • Tizen
  • webOS
  • Lightning.js
Back-end
  • Node.js
  • NestJS
  • Go services
  • Rust services (Axum, Tokio)
  • FastAPI
  • GraphQL
  • gRPC
  • REST
Databases
  • PostgreSQL
  • MySQL
  • MongoDB
  • Redis
  • ClickHouse
  • Elasticsearch
  • pgvector
  • Qdrant
Messaging and events
  • Apache Kafka
  • RabbitMQ
  • NATS
  • Redis Streams
  • Amazon SQS
Data and analytics
  • ClickHouse
  • Matomo
  • dbt
  • Apache Airflow
  • Metabase
  • Grafana
AI
  • Claude
  • OpenAI
  • Gemini
  • Llama and open models
  • Vercel AI SDK
  • LangGraph
  • Hugging Face
  • vLLM
Automation
  • n8n
  • Temporal
  • Zapier
  • Make
  • Custom workflow engines
Video
  • FFmpeg
  • HLS
  • MPEG-DASH
  • Widevine
  • FairPlay
  • PlayReady
  • Shaka Player
  • ExoPlayer
Cloud and DevOps
  • AWS
  • Google Cloud
  • Azure
  • Cloudflare
  • Docker
  • Kubernetes
  • Terraform
  • GitHub Actions
  • Prometheus
  • Grafana
Quality
  • Playwright
  • Cypress
  • Appium
  • Detox
  • k6
  • OWASP ZAP
  • BrowserStack

Ways to work with us

Pick the shape that fits your organisation. All four come with the same senior people and the same standards.

Project delivery

A defined scope, a delivery plan and a team that owns the outcome end to end.

Dedicated team

A long-running squad embedded with your product organisation, managed by us, directed by you.

Team extension

Senior engineers, designers or QA specialists who join your existing team and process.

Support and operations

Ongoing maintenance, monitoring and on-call for software we built or inherited.

Not sure which service you need?

Describe the problem and we will tell you honestly what it takes, including when the answer is a smaller project than you expected.