# Meshkloud

:::info
#### Sprint Meeting
#### **Date:** 11:00 AM - 11:40 AM EDT 04/16/2026
:::

**Meeting URL**: https://us05web.zoom.us/j/85185421107?pwd=cEMVqz9XbDzI1E56aG17B7nnHKOGnN.1&from=addon
**Meeting ID**: 851 8542 1107
**Passcode**: 2Y88Vk

## What I did
Over the past two weeks, I focused mainly on improving the authentication experience, expanding backend support for pipeline and timesheet workflows, strengthening shared frontend components through PR reviews, and supporting team coordination.

> **Authentication and onboarding**
> I reviewed the existing auth flow through Figma and the codebase, and separately analyzed the new sign-in and sign-up flow as it evolved. I’m fully aligned with the updated direction.
***(please review the below new sign-in/sign-up flow)***

> **Pipeline, proposals, and invites**
> A major part of the work was around the pipeline area. I built endpoints for the Pipeline List and worked on the pipeline summary APIs. I also continued backend work for proposals and invites flow APIs, helping move the pipeline-related functionality closer to a more complete and usable state. Alongside that, I reviewed frontend work for the Pipeline page to make sure the UI and backend direction stayed aligned.

> **Timesheet functionality**
> On the Timesheet side, I worked on APIs for setting and fetching employee roles, which supports role-based behavior and data handling on that page. I also continued work directly on the Timesheet page as part of the broader feature development effort.

> **Email and system integration**
> I worked on the problem-awareness email template based on the new email marketing design.

> **Frontend components and code review**
> A significant portion of the last two weeks also went into code review and UI quality control. I reviewed, commented on, and merged multiple PRs related to shared and global components, including button, badge, text input, modal, checkbox, radio, toggle, spinner, notification, and change request modal components. I also re-reviewed updated PRs after feedback and merged the ones that were ready. This helped keep component quality consistent across the product.

> **Team coordination and support**
> Beyond direct implementation work, I also supported progress across the team by assigning Muhammad a task for the simplified button-additional component and setting priority for Valeriy’s global component task. I reviewed Valeriy’s PR as well and merged it after verification. This helped keep parallel workstreams moving while maintaining review quality.

**Overall outcome**
Overall, the last two weeks were focused on moving core product areas forward in a practical way: improving auth flows, building backend support for pipeline and timesheet features, refining shared UI components, and helping the team stay aligned through review and coordination.

## New Sign-in/Sign-up flow
> I reviewed the updated sign-in and sign-up flow, and overall I agree with the direction. I think the strongest part of this approach is that it is based on user intent instead of forcing people to define their role too early. So rather than asking users something like who they are, the flow guides them based on what they want to do, like create a workspace, join a project, list a firm, find work, or just explore the platform. I think that makes the onboarding much more natural and avoids unnecessary confusion for different user types.
>
> From the user experience side, the structure is also much clearer now. All users start from the same shared foundation with the welcome step, email verification, and password creation. After that, the flow branches depending on the user’s goal. Self-service users go through the start selection, while invitation-based users skip that step and go directly into the invitation path, which makes sense because their destination is already defined.
>
> For invited users especially, I think the branded invitation flow is a strong idea. Showing the inviting company, the project name, and the assigned role right away gives users immediate context about why they are there and what they are joining. That should make the experience feel more trustworthy and much easier to understand, especially for clients, consultants, subcontractors, or team members joining through an invite.
>
> One specific improvement we discussed is the invite experience. Volo suggested replacing the invite code step with a magic link, and I agree with that. From a product and UX perspective, the magic link approach feels much cleaner. It removes the extra manual step, reduces friction, and lets the user land directly in the correct onboarding path with the right project and access context already attached. I think that is a better experience and a natural improvement for the invitation flow.
>
> So overall, my view is that the updated flow is going in the right direction. It is simpler, more contextual, and more scalable for different kinds of users. And with the magic link replacing the invite code section, the onboarding experience becomes even smoother and more intuitive for invitation-based users.

## Blockers
There is a slight delay on the frontend design side, but it’s not a major blocker. It would still be helpful to have the full design completed within the next 3–4 weeks to keep development moving smoothly.

## Expected timeline
**Sign-in/sign-up flow**
The design team is currently working on the authentication screens. In parallel, I will proceed with the backend logic. This part is expected to take around ***1 week***, including development and testing.

**Project List**
The project list feature (frontend + backend, including API integration) is estimated to take around ***1 week*** once everything is aligned.

**Timesheet**
Muhammad is currently working on frontend UI adjustments and is expected to complete them today. From tomorrow, I will start API integration, which should take about ***1 week***. We’ll also need the design for the personal tab to be finalized to complete this part smoothly.

**Pipeline**
The pipeline feature (frontend + backend, including API integration) is estimated to take around ***2 weeks***. It would be important to have the design finalized before or during this period to avoid delays.

## OCR project
> Chris shared an idea for building an OCR-based PDF processing tool that goes beyond standard text extraction and moves toward AI-driven document understanding. The proposed concept includes extracting file names, sheet numbers, and version numbers, with future potential to capture more structured data such as room names and areas. He also mentioned features like splitting multi-page PDFs into individual sheets, auto-renaming them, comparing drawing versions, and highlighting changes. The long-term vision is for the system to improve over time by learning from the team’s typical plan sets.
>
> I responded positively to the idea and agreed that the strongest value is in understanding and structuring the data, not just performing basic OCR. He noted that the right technical approach will depend on the complexity and consistency of the sample PDFs. For simpler files, he believes an OCR plus AI API approach could move quickly. For more complex or inconsistent documents, he expects additional custom logic or a more tailored model may be needed over time.
>
> Overall, I sees this as a strong direction. He asked for sample PDFs to better evaluate complexity and confirmed that, once Chris has a prototype ready, he can review it, refine it, and handle integration if it fits the product workflow.