Web apps and integrations
Software development
Custom web apps, internal tools, and connections between the systems you already use, built clear enough to maintain after we hand them over.
We build software and AI your team will actually use.
Custom apps, AI chatbots, and AI agents, connected to the systems you already run, without replacing what works.
Connected through MCP: an open standard for linking AI to your tools, with the access you choose.
You already have the systems.
Illustrative examples
A CRM, an inbox, spreadsheets, a project board, accounting, a helpdesk, an internal app, a database that has been running for years. That stack is not the problem to solve — it is the foundation to build on.
- CRM Example · connected
- Inbox Example · connected
- Spreadsheets Example · connected
- Project board Example · connected
- Accounting Example · connected
- Helpdesk Example · connected
- Internal app Example · connected
- Legacy database Example · kept manual
We don't replace your stack. We connect it.
Software. Chatbots. Agents.
Three things we build
Answers from your data
AI chatbots
Chatbots for your customers or your team that answer from your own documents and records, limited to what each person is allowed to see.
Routine work, with approvals
AI agents
Agents that carry out routine, multi-step work across your tools, and wait for a person to approve the steps that matter.
Built by us: Tamoogi.
Our own product
Tamoogi is our own product: a workspace where Philippine business owners track the permits, filings, tasks, and expenses of every business they run. It brings all three of our services together, and it is live at tamoogi.com.
In Tamoogi
Software development
A full web app with subscriptions and card payments through PayMongo, separate permissions for each collaborator, private file storage, and a complete data export.
In Tamoogi
AI chatbot
An assistant that answers from each business’s own records, only those the user is allowed to see, and removes card numbers and tax IDs from the records it looks up before they reach the AI model. Task lists in its replies are read from the records themselves, so it cannot list a task that does not exist.
In Tamoogi
AI agent
It creates tasks, logs expenses, reads receipt photos, and sets reminders when asked, and waits for a person to press Confirm before any money change or deletion.
In Tamoogi
Connected through MCP
Its own MCP server lets people use Tamoogi from Claude, ChatGPT, and code editors, with more than 60 tools for tasks, expenses, documents, and reminders. Each person signs in with their own account, and every tool follows the same permissions as the app.
One workflow, across the systems you already use.
Illustrative example — not a client system.
- Inbound A lead lands in your inbox.
- Route The AI reads it and decides where it goes.
- Draft The AI layer drafts a reply.
- Check A person approves or edits it.
- Write The result is written back to your CRM.
Automate the routine, not the reasoning.
Bring us a workflowWhere AI does not belong.
Judgment stays with people
Some steps should stay with your people. We mark them kept manual, and build the automation around them rather than through them.
- Approving spend Human judgment stays here Kept manual
- Performance conversations Human judgment stays here Kept manual
- Legal and contractual commitments Human judgment stays here Kept manual
- The final send Human judgment stays here Kept manual
Start with the work. Then build what helps.
How we work
-
Map the current work
We begin with the decisions, handoffs, and data that shape the work today.
-
Find the smallest helpful step
The next useful move is usually smaller than a rebuild. We find it and prove it.
-
Build it into your stack
Standalone tools and integrations that fit how your team already works.
-
Keep it maintainable
Systems clear enough to maintain, adapt, and trust long after we hand them over.
Two people who stay close to the work.
A two-person studio
A two-person studio pairing cloud and AI engineering with integrations and delivery. The people who map your workflow are the people who build it.
Rebie Estares
Cloud, AI & software systems
Builds Azure-based software and AI tools that make everyday work easier for the people using them.
- Microsoft Certified Professional
- MCSD: App Builder · MCSA: Web Applications
Valerie Estares
Integrations, full-stack & delivery
Designs secure integrations and helps teams deliver software they can understand, maintain, and improve.
- Master of Innovation and Business · Asian Institute of Management
- Lean Six Sigma Green Belt
Latest insights.
All insights →
What we learned connecting a real app to AI through MCP
Lessons from giving AI access to a live business app through MCP, covering permissions, approvals, accurate figures, and protecting sensitive data.
Oct 13, 2026
How to Audit a Business Process Before Automating It
How to audit a business process before automating it, with a practical 5-step framework to find bottlenecks, measure ROI, and pick what to automate.
Sep 23, 2026
Learning marketing from a technical starting point
Two technical people share how they are learning marketing by building a thoughtful, AI-assisted process around real conversations and useful work.
Sep 9, 2026Tell us what needs untangling.
Whether you need a new app, a chatbot, an AI agent, or AI added to a system you already run, tell us what you run and where it hurts.