Evidence before commitment. The smallest solution that works.
We'd rather scope a narrow, honest project than sell a big one we can't back up. Here's how we actually work, from the first conversation to ongoing support.
Start a ConversationHow a project actually moves.
Listen
We start with the people living with the problem, not an assumption about what they need.
Investigate
We review your existing systems and constraints closely enough to scope real work, not a guess.
Scope
A narrow, honest proposal: the smallest thing that actually solves the problem, its timeline and its price.
Build & Launch
We build directly with you, with clear deliverables and checkpoints along the way, not a black box until the end.
Support & Learn
Where it's part of the agreement, we stay involved: monitoring, fixing, and improving what we shipped.
What we hold to on every engagement.
Evidence before big commitments
We review your actual systems and talk to the people living with the problem before proposing anything substantial.
The smallest responsible solution
We scope narrowly: the smallest thing that actually solves the problem, with its own timeline, risks and price, not a maximal rebuild nobody asked for.
Direct technical collaboration
You work directly with the person building it. We'll name the lead once one is assigned; not every Fellow in the community is staffed on your project.
Implementation & handover, not an afterthought
Deployment, training, documentation and support are part of the work when they're part of the agreement, not something we figure out after launch.
Security and privacy by default
Access control and data minimization are part of how we build, not a checkbox at the end. If a requirement, like a specific compliance certification, is outside what we currently support, we'll say so plainly.
What we can do today, what needs a partner, and what we don't offer.
Can do today
- Scoped software, data and AI projects with a clear deliverable
- Integration work between systems you already run
- AI-assisted features with a human-approval step built in
- Implementation, training and ongoing support where contracted
Needs a partner
- Specialized regulatory or compliance audits outside our current certifications
- Large-scale enterprise data-warehouse migrations
- Work requiring on-site staff in a location we don't currently operate in
Not offered right now
- Fully autonomous AI systems with no human review on consequential decisions
- Guaranteed 24/7 uptime commitments without a contracted support agreement
- Any engagement that skips a scoping conversation first
From first conversation to ongoing support.
- 01Introductory conversation
- 02Problem review
- 03Scoping proposal
- 04Contract
- 05Work
- 06Acceptance
- 07Support
Have a problem that keeps coming back?
Tell us what's happening, what's been tried, and what a better outcome would look like.