Technologies
The stack we work in, and work inside.
Six areas, one team. This is what our engineers use daily — not a list of everything we have heard of, and not a recommendation for what you should be running.
Areas covered in house, by one team
The centre of gravity, and has been from the start
We work in your stack, not ours — no migration pitch
Reseller deals. Nothing here earns us a referral fee
Grouped by the problem it solves.
A stack list sorted alphabetically tells you nothing. These are grouped the way the work actually arrives.
AI & LLM engineering
The systems around a model, which is where nearly every AI feature actually succeeds or fails: retrieval, tool use, evaluation and the cost of serving it.
- Large Language Models (LLMs)
- RAG
- LangChain
- LangGraph
- AI Agents
- MCP (Model Context Protocol)
- Vector Databases
- LlamaIndex
- Embeddings & semantic search
- Reranking
- Prompt engineering
- Function & tool calling
- Structured output
- Hugging Face
- OpenAI API
- Anthropic Claude API
- Ollama
- vLLM
- Self-hosted inference
- Evaluation harnesses
Machine learning & data science
Models trained, evaluated and shipped — with the measurement in place before the improvement, so a change can be argued about with a number rather than an opinion.
- Machine Learning
- Deep Learning
- PyTorch
- Scikit-learn
- NLP
- Computer Vision
- YOLO
- OpenCV
- Transformers
- TensorFlow
- Keras
- XGBoost
- MLflow
- Model evaluation
- Feature engineering
- Fine-tuning
- ONNX
- Time series forecasting
- Recommendation systems
- MLOps
Backend & APIs
Services that hold up under real traffic, with contracts a client engineer can work against without having to ask what an error means.
- Python
- Django
- Django REST Framework (DRF)
- FastAPI
- Flask
- JavaScript
- TypeScript
- Node.js
- REST APIs
- GraphQL
- WebSockets
- Celery
- Pydantic
- SQLAlchemy
- Authentication & OAuth
- Payment integrations
- Background jobs
- Next.js
- React
- API design
Databases & data modelling
Schemas that survive the second year, queries that stay fast as rows accumulate, and migrations that can run against live traffic.
- PostgreSQL
- MySQL
- SQL Server
- MongoDB
- Database Design
- SQL
- Redis
- Supabase
- pgvector
- Pinecone
- Qdrant
- Chroma
- Query optimisation
- Indexing
- Schema migrations
- Data modelling
- Elasticsearch
- Caching strategies
Cloud, DevOps & infrastructure
Deploys that are dull on purpose, infrastructure written down as code, and a cloud bill somebody actually owns.
- AWS
- GCP
- Docker
- CI/CD Pipelines
- GitHub Actions
- Nginx
- Linux
- Git
- GitHub
- Kubernetes
- Docker Compose
- Vercel
- Terraform
- Load balancing
- SSL & domains
- Monitoring & logging
- Cloud cost optimisation
- Serverless functions
Data, analytics & automation
Pipelines and reporting that a business can trust, plus the automation that stops work being copied between systems by a person.
- Pandas
- NumPy
- Power BI
- Excel
- Matplotlib
- Seaborn
- Plotly
- n8n
- ETL pipelines
- Data cleaning
- Dashboards
- Web scraping
- Airflow
- Workflow automation
- Zapier & Make
- API integrations
- Reporting automation
- Jupyter
A stack list is not the same as judgement.
Any agency can print this page. What decides whether a project lands is knowing which of these tools the problem actually needs — and being willing to say when the answer is a smaller one than you were expecting.
We have told clients their retrieval problem was a chunking problem, and their model problem was a data problem. That conversation is free, and it happens on the first call.
The engineers behind the stack.
Each discipline has its own bench and its own published rate — and the same engineers build the projects we take on end to end.
AI/ML engineers
Engineers who have taken an LLM feature past the demo — through evaluation, latency, cost and the failure modes that only appear with real users.
Backend engineers
Service and API engineers who have carried a pager, and who write the boring, well-tested code that keeps you off one.
Data engineers
Pipeline and warehouse engineers who care whether the number in the dashboard is correct, and can prove it.
Cloud & DevOps engineers
Infrastructure engineers who reduce the number of things that can page you, and who treat your cloud bill as an engineering artefact.
Or see some of what we have built with this stack, all of them live and public.
Questions about the stack.
Do we have to use your preferred stack?
No. We work inside whatever you already run. Nothing on this page is a recommendation to migrate, and we have no reseller relationship with any vendor on it — so we gain nothing from moving you.
Something we need isn't listed. Does that rule you out?
Usually not. Most of what matters transfers: a senior engineer who has shipped Django services will be productive in Rails or Laravel within a fortnight. Ask on the call and we will tell you honestly whether it is a fortnight or a bad idea.
Which of these do you use most?
Python and the AI stack around it, day to day. Most of what reaches us is either an LLM feature that has to survive real users, a backend that has outgrown its schema, or a data pipeline nobody trusts any more.
Can you take over something already built in these?
Yes, and a good share of our work is exactly that. We read what exists and match its conventions before proposing changes to them. We do not open with a rewrite recommendation.
Tell us what you're running.
Thirty minutes with an engineer who works in this stack daily. You will get a view on approach whether or not you hire us.