I'm a product designer in Toronto. I've redesigned a 16-year-old medical records system, designed AI document processing at OpenText, and rebuilt booking and payments for a global spiritual platform.
A 16-year-old medical records system, redesigned
One site for every city, and a front door for temples
Rituals booked online, for a global audience
Keeping key details in view as documents move
I studied computer science and software engineering before moving into design, so I think about how a screen will be built while I design it. That keeps my work feasible and makes handoff to developers smooth.
Since 2023 I've also led client research in financial services, designed training programs, and now direct a 20-person team at Bhakti Academy Toronto, where I'm also leading our new multi-city website. Explaining complex products to real people is what I do best, and I'm now returning to product design full time.
Outside work: poetry, meditation and running.
A 16-year-old medical records system, redesigned
Clinic and hospital staff relied on an EMR whose navigation changed from screen to screen and whose interface looked its age. I led design on a team rebuilding it to be consistent, responsive and accessible.
Draft for review: "What I'd do differently" is my suggestion. Rewrite it in your own words.
Inconsistent navigation, hard-to-read screens and user complaints across a 1,200+ screen product.
Design team lead and UI/UX designer. Research, information architecture, interaction design, testing.
Design, development and business analysis. Client under NDA.
average satisfaction in user testing.
Tools Adobe XD, Maze
I interviewed four clinic and hospital staff over Zoom and turned what I heard into two personas. Their frustrations set the priorities for the redesign.
Needs to return patient calls and leave on time, and to keep the practice running smoothly.
Frustrated that the fonts are hard to read, and there is too much to enter while patients wait.
Needs a paperless practice within two years and clear, quick communication with patients.
Frustrated by alert fatigue, and by spending 2 to 4 hours a day after work finishing tasks.
The front desk's most common job is finding a patient and booking them. I made the scheduler the home screen and put patient search one step away, with selecting or adding a patient leading straight to booking or demographics.
"The fonts are hard to read" came up directly in research, and AODA compliance was a project goal. So contrast, type size, labels, hint text and error messages went into the components themselves, and every screen built from them inherits the fix.
With 1,200+ screens across 25+ modules, consistency could not depend on memory. I maintained the style guide and icon set, built screens from shared components in Axure, and wrote Jira user stories so developers built from the same parts.
I explored layouts on paper before building clickable prototypes. Below, an early sketch of a collapsible list, and the prototype for the lab reports module.
We ran A/B and usability tests with clinic and hospital staff on Zoom and Maze, read heatmaps to see where attention went, and ran heuristic evaluations before handoff.
Flows took fewer clicks, the information architecture was rearranged so navigation made sense, and icons, button labels, hint text and error messages were rewritten to be clear.
Listening is the most important skill to have in a meeting.
Leading design across three disciplines taught me that the best decisions came from hearing developers and analysts out before proposing anything. What I'd do differently: run a larger round of testing before high fidelity. Four interviews and six testers told us a lot, but more clinicians earlier would have caught issues sooner.
References from this project are available on request.
Keeping key details in view as documents move
OpenText Core Capture uses AI to classify scanned documents and pull out their data. As documents moved through the workflow, the people reviewing them kept losing track of critical information tied to specific parts of a page. I designed the features that fix that.
Waiting on Vishesh: did these features ship? Screens are from the prototype and use sample data.
Critical details tied to one part of a document got lost as it moved from stage to stage.
UX designer. Research, flows, prototyping, testing and stakeholder reviews.
Axure RP with OpenText's design system, Figma, Zoom for remote research.
features taken from research to stakeholder approval.
I interviewed users over Zoom, studied competing capture tools, and grouped the findings with affinity diagrams and journey maps.
The problem was information tied to one part of a page, so annotations sit on that part. Reviewers can redact, stamp, highlight, add text, draw an arrow or comment, right where the detail is.
The workflow runs from scan and import to organize, review and submit. I made annotation reachable from both Organize and Review, the stages where people spot problems and hand work on.
I designed with the company's existing components rather than inventing new ones. It sped up prototyping, kept the feature consistent with the rest of the product, and made approval and handoff easier.
The first version of the review screen was a pen sketch: documents across the top, the open page below, annotations on the side. From there I built high-fidelity clickable prototypes in Axure, ran remote A/B and usability tests, and refined the design in stakeholder reviews. Live testing was planned for OpenText World.
A well-defined design system saves time and keeps applications consistent.
Working inside a large company's system taught me to spend my effort on the problem, not on reinventing buttons. What I'd do differently: measure how long it takes reviewers to find a flagged detail before and after, so the impact is a number, not just a good feeling.
References from this project are available on request.
Rituals booked online, for a global audience
The Linga Bhairavi website buried its content, did not work on phones, and most of its rituals could not be booked online. I led UX for a rebuild that now lets people around the world book and pay for rituals from home.
Waiting on Vishesh: a screenshot of the old site for a before-and-after.
Information overload, no mobile support, and most rituals only bookable in person.
Team lead and UX designer. Research, flows, wireframes, prototypes and testing.
Modernize the site, grow paid services, and reach more people.
rituals now bookable and payable online.
Tools Figma
I interviewed seven visitors in person at meditation halls, found four groups of users, and focused on the two the site served least well.
Wants meditations centred on the feminine, and to learn about Eastern traditions.
Struggles to find authentic stories and people who share her interests.
Wants to book rituals for his family from home, and to know which days are special.
Frustrated by visiting temple offices and filling in forms every time he books.
People like Vinay book around auspicious lunar dates, so the calendar is the heart of booking: highlighted dates, then a time slot, then in person or online. I tried several versions of the calendar and form before settling on one.
The old site did not work on phones or tablets. I took the navigation from a paper sketch to a wireframe to the final visual design, with each section given a clear image and label.
With dozens of rituals, people needed to compare them at a glance. I explored card layouts and chose one that pairs an image with the essentials and a clear booking button.
Payment is where people drop out. I cut the flow to the information people need to decide (date, time, mode and total cost) and nothing else.
Wireframes were built in Figma and handed to the visual design team. We ran A/B and usability tests in person, plus heuristic evaluations, before launch.
Less is more in payment flows.
We were lucky to have easy access to users, and it showed: every big decision traces back to something we heard at the meditation halls. What I'd do differently: track bookings and drop-off from launch day, so the redesign's impact on revenue could be measured, not just seen.
References from this project are available on request.
One site for every city, and a front door for temples
Bhakti Academy runs live-in programs in Toronto and Montreal, with more cities planned. Its website was built for one city, students couldn't find dates or costs, and the support office that helps temples start new academies had no home online. I'm leading a single master site that serves every city.
Draft for review: "What I've learned so far" is my suggestion. Rewrite it in your own words.
A one-city site with no dates or costs, nothing for temples, and a blog search engines can't read.
Project manager and design lead: information architecture, user flows, key design decisions, branding. I sign off design.
Developers (detailed UI and build), city teams, and the BAOGF support office. Weekly reviews.
launch target. Structure and flows done; wireframes in review.
Tools Claude (AI) for concept wireframes and diagrams, Figma
The developers wanted to start building quickly, and the launch date is fixed. Rather than run a full research program, I focused on the parts that are hardest to change later: the site map, the navigation, and the paths to the two most important actions, applying and enquiring.
I drew on my own experience as a former student and now Program Director, feedback gathered from students, staff and temples, and an audit of the current site.
Today the site splits by city only at "Join". I structured it as shared content written once, one template every city fills in, and a separate section for temples. Visitors pick a city early, so dates, costs and contacts are always local, and a new city needs content, not a redesign.
Each city relies on its own application form for its back-end data, so replacing the forms wasn't an option. Instead, the site prepares people before the form (dates, cost, eligibility, common questions) and welcomes them back after it with clear next steps.
Temples are a different audience, closer to business clients than students. BAOGF gets its own entry from the home page, and its pages answer the four questions temples ask first before asking them to enquire.
Instagram and WhatsApp carry most of the conversation today. A global blog of stories from every city, tagged by city and readable by search engines, gives the site its own way to bring people in. Every post ends with "Apply in your city".
I used Claude to turn the site map into concept wireframes for the key screens, desktop and mobile, in days rather than weeks. I then reviewed and refined them in Figma, with a note on each decision, so the developers could start building sooner. Tap any one to enlarge it.
When the timeline forces shortcuts, get the structure right.
Pages and colours are easy to change after launch. The site map, the navigation and the paths to applying are not. Leading the project as both manager and designer has also taught me to make design decisions the team can act on the same week.
References from this project are available on request.