No partner badges here. This is the list of models, clouds, databases and business platforms our engineers use in production, with the reasons we pick one over another.
A platform appears here because our team has shipped production work on it. It does not mean we hold a partner tier, a certification or a reseller agreement with that company, and the names are trademarks of their owners. Where we do hold formal partner status we will say so explicitly. When a recommendation would also benefit us, for example proposing our own Eventify platform, we tell you.
Every model sits behind one interface in our applications, with routing, evaluation and guardrails in our code. Changing the model is a configuration change with an evaluation run, not a rewrite.
Anthropic Claude
Our usual first choice for agents, tool use and long-context work. We build with the Claude API, the Model Context Protocol and Claude Code.
The AI layer is the one that changes most often, so it is the one with the clearest boundary. Models are chosen per workload and can be replaced after an evaluation run.
Business platforms, cloud and data
The rest of the stack
Salesforce
Implementation, Apex and Lightning Web Components, Flow, integrations and Agentforce. A dedicated practice with its own managed services.
Platform choice is made in the discovery sprint, written into the statement of work, and revisited only with an evaluation to justify it. These are the rules of thumb we start from.
If your situation is
We usually start with
Because
Agents, tool use, long documents
Anthropic Claude
Strong tool use and instruction following; MCP fits our tool layers.
Data must not leave your network
Open-weight models in your VPC or on-prem
Full control of data and cost; we handle hosting and evaluation.
Already on Google Cloud or Workspace
Google Gemini
Identity, data and billing are already in place.
AI inside the CRM for sales or service
Salesforce Agentforce
Keeps context, permissions and audit inside Salesforce.
Vector search, moderate scale
PostgreSQL with pgvector
One database to operate; hybrid search with full-text in the same query.
Vector search, large scale or high QPS
Milvus
Purpose-built indexing and scaling.
Static sites, edge APIs, small services
Cloudflare Workers, Pages, D1, R2
Low operational overhead and global edge delivery.
Payments and subscriptions
Stripe
Mature APIs, webhooks we already handle idempotently.
Are you an official partner of Anthropic, Salesforce or AWS?
This page lists platforms our engineers have shipped production work on. It makes no claim about partner tiers, certifications or reseller status. Where we hold a formal partner status we will state it explicitly on this page.
Can you work with the provider contracts we already have?
Yes. Most clients bring an existing cloud account, model agreement or Salesforce org. We build inside your accounts so that you keep the data, the billing relationship and the ability to change supplier.
What if our stack is not on this list?
The list is where we spend most of our time, not a boundary. We regularly work with other databases, clouds and SaaS platforms through their APIs. Describe your environment on the contact page and we will tell you honestly whether we are a fit.
Do you recommend one language model over the others?
We recommend a model per workload, behind an interface that lets you change it. Anthropic Claude is our usual first choice for agentic and long-context work, open-weight models where data must stay inside your network, and Gemini or OpenAI where your ecosystem already favours them. See on-prem versus cloud LLMs.
Ready to make AI work in production?
Tell us about your systems, your data and your constraints. We reply within one business day with a straight answer and a suggested next step.