About Rawleaf
Experienced people who use AI every day.
Rawleaf pairs years of senior technology experience with hands-on, daily practice. We don't just advise on AI: we run our own work on it, build products with it, and know where it helps and where it doesn't.
AI agents, in daily practice
We run our own business on specialist AI agents every day, for inbox triage, meeting notes, contract preparation, hiring, compliance, finance and coding. So anything we recommend to you has been tried on our own work first.
Agents that work like a team, not a chatbot
We break a role into focused agents with clear responsibilities, standing rules and handoffs between them, so each one does one job well.
Rules that hold
We write durable operating rules for every agent: what needs a person's action, what's just for information, and what must never mix. Then we refine them as the work changes.
Meetings into actions
Turning meeting transcripts into decisions, actions and tracked follow-ups across suppliers and internal teams.
Connecting AI to the systems you already use
AI is only useful when it can reach your real work. We connect it to the tools businesses already rely on, using standard connectors and APIs.
Your everyday tools
Gmail and Google Workspace, calendars, Slack and Google Chat, Notion, Jira and Confluence, Linear, Miro, GitHub, AWS, meeting-note tools and WhatsApp messaging.
Automations that run themselves
Scheduled and event-driven automations: daily summaries, inbox triage, meeting follow-ups, customer messages, monitoring and alerts.
Reusable building blocks
We wrap your systems' APIs into reusable skills and plugins, so new automations are quicker to add and easier to maintain.
Building with AI
We use AI coding agents to take products from specification to working software, quickly and properly. Fitovo, our adaptive training app, is live on the App Store and was built this way.
Decision support with the right controls
Where AI helps people make decisions, we design it so a person makes the final call, can always override it, and its accuracy is measured over time.
Removing friction, not just effort
The best automations don't only save minutes. They remove the 'who do I ask?' and 'which form is it?' moments that slow everyone down.
Strategy, governance and security
More than 20 years running technology in healthcare and transport, where mistakes are expensive. We look at AI the way you'd want a careful senior adviser to: what it's worth, what could go wrong, and what it will cost to run.
Regulated data, designed in from the start
Data protection impact assessments where they're needed, careful handling of sensitive personal data, and keeping personal data away from any system that doesn't need it.
Security and compliance that's practical
Experience with information security management systems, payment card security (PCI DSS), UK GDPR, staff phishing awareness and keeping cloud costs under control.
AI-assisted security testing
Using AI tools to help test the defences of live systems, so problems are found by you first.
The right tool for the job
Matching the AI model and where it runs to the workload, on cost as well as capability. Sometimes the right answer is a simpler, cheaper option.
How we keep you in control
- You decide where the line is. Routine replies in wording you've approved can go out automatically.
- Anything that commits you waits for a person: nothing is priced, paid, signed, deleted or bought without approval.
- Everything the assistant does is logged, so you can see what it did and why.
- Data from different businesses is always kept separate.
- We tell you where AI won't help, as well as where it will.
What we don't do
We're specialists in applied AI: putting existing AI models to work safely in real businesses. We don't train new AI models from scratch or do academic machine-learning research, and we'll say so if that's what your project needs.
See what we've built