Tama-OS

Tama OS is a local-first deterministic decision-readiness layer that tells a personal investor what evidence or financial safeguard is missing before they take more risk.

Comment

Inspiration

I built Tama OS because I needed a system for the way I actually make decisions. I am easily distracted by new ideas, sometimes impulsive, and I love collecting data. I also tend to overengineer my own systems. Before Tama OS, I had two separate HTML-based tools: one for personal finance and one for investment research. They helped, but they were still separate apps. I could track what happened, but I still had to manually connect my money records, portfolio, research notes, and next decision. So I started discussing the bigger direction with ChatGPT: what if this was not just a finance tracker, but a local-first financial operating system? Because my coding skill is limited, I acted more like the product brain. I brought the problem, workflow, constraints, and direction. Codex helped as the coding partner that turned those ideas into working HTML, CSS, JavaScript, tests, and documentation. Tama OS came from that collaboration. Its goal is simple: help me pause, understand what is missing, and answer, “What should I do next?”

What it does

Tama OS is a local-first AI Financial Operating System. It connects a Money Workspace and Research Workspace through a read-only Decision Hub. The Decision Hub reads local browser data, runs deterministic checks, and creates Today’s Brief. It can detect issues like stale records, missing data, or an investment holding without a decision-ready research thesis. GPT-5.6 is used as a Copilot to explain the deterministic recommendation in plain language. It does not replace the rules, mutate user data, trade, or provide financial advice. If no API key is available, Tama OS still works in offline deterministic mode.

How we built it

Tama OS was built with HTML, CSS, vanilla JavaScript, localStorage, Codex, ChatGPT, and GPT-5.6 through the OpenAI Responses API. Before Build Week, I had two standalone browser apps: one for finance and one for research. During Build Week, I used Codex to help inspect the existing code and add a new read-only Decision Hub on top instead of rewriting everything. The Build Week work added local snapshot helpers, a deterministic decision engine, Today’s Brief, GPT-5.6 Copilot support, fictional demo data, shared navigation, tests, and documentation.

Challenges we ran into

The biggest challenge was combining two existing apps without breaking them. They had different state, storage, and internal structure, so a full merge would have been risky. Another challenge was my limited coding skill. I had to be clear about the product direction and use Codex as a real coding partner. The finance context also required clear safety boundaries. GPT-5.6 only explains deterministic findings. It does not make hidden decisions, mutate data, or act as a financial advisor.

Accomplishments that we're proud of

I am proud that Tama OS became more than a tracker. It now turns scattered money and research records into one practical next action. I am also proud that the existing Finance and Research apps were preserved instead of replaced. The new OS layer adds value while keeping the original tools usable. Personally, I am proud that I could turn a messy personal workflow into a working product by combining product thinking, ChatGPT, Codex, and a local-first architecture.

What we learned

I learned that the hardest part of personal finance is not always tracking data. It is turning data into a decision. I also learned that AI works better with clear boundaries. In Tama OS, deterministic rules create the recommendation, and GPT-5.6 explains it. Most importantly, I learned that someone with limited coding skill can still build a meaningful product by using AI tools carefully and staying focused on the real user problem.

What's next for Tama-OS

Next, I want to expand the deterministic decision engine, improve Today’s Brief, add Financial Memory for user rules and goals, and create a decision journal. Long term, Tama OS can grow from a Financial Operating System into a broader Personal Operating System for career, learning, reading, health, and projects.

Built With

Share this project:

Updates