Aqua X · 2022 to 2024

Cyber Dive · Aqua XProduct Design Lead2022 to 2024

Recovery center operations, rewritten for residents

Aqua X is the smartphone built for recovery centers. A purpose-built device paired with a staff oversight portal, designed around tiered access that adapts as residents progress through their program.

Metric

alert triage resolution time: 24hrs → 15min
active deployments across recovery centers: Multi-state
resident launcher + staff portal + workflow: Full-stack

Product Design Lead · Cyber Dive · Aqua X

Partial case study, details under NDA

Aqua X · Product Design

Aqua X: Recovery center operations, rewritten for residents

2.5 years, Cyber Dive (multiple releases)

02 The 30-second version

The 30-second version

The challenge

Recovery centers had three bad choices for resident phones: ban them (residents feel isolated, sneak in hidden devices, drop out of programs early), allow them freely (old contacts, relapse triggers, no accountability), or repurpose parental controls built for teens (easy to bypass, no progression model, wrong mental model entirely). None of these supported recovery as a gradient.

What I did

Designed a purpose-built Android launcher on a Pixel device paired with a staff web portal. Residents receive a phone on day one with a tier appropriate to their program phase. Staff assign permissions per tier (contacts, apps, internet, location, camera), promote residents through tiers as they progress, and monitor activity via an instant-replay feed instead of live surveillance. The phone became a tool for reentry, not a tool for control.

My role

Product Design Lead, Cyber Dive
Partnered on qualitative research with recovery center admins to map existing staff workflows end-to-end
Contributed to mental-model interviews with residents to surface pain points around phone access, reentry, and program progression
Owned the resident-facing Android launcher and the staff oversight portal (web)
2.5-year engagement across research, design system, and multiple shipped releases

The team

Anirudh PalaskarProduct Design Lead
Cyber Dive Product TeamEngineering, Research, Operations
Partner Recovery CentersDesign Partners and Early Users

The key insight
The category was stuck on a binary: phones on, phones off. But recovery itself is not binary. It is a gradient from high-risk to self-sufficient, with specific program phases mapped to specific resident capabilities. The product had to mirror that gradient. Tiered access was not a feature. It was the mental model that made every other decision fall into place. Once the team stopped asking "what permissions do residents get" and started asking "what phase is this resident in," the rest of the design followed.

The constraints

Progression, not permissionRecovery happens in phases. Access had to adapt as residents moved through the program, not stay fixed at a single permission level.
Accountability without confiscationStaff needed to see what was happening without taking phones away. Taking phones away was the failure mode the old category was built on.
Non-technical staff as primary operatorsCounselors and program directors were the daily operators, not IT admins. Every interaction had to be legible to someone who has never configured an MDM.
Early intervention over after-the-fact reportingThe product had to surface risk signals while there was still time to act, not generate compliance reports after a crisis had already unfolded.

03 Key decisions

Key decisions

01
Tiered access, not all-or-nothing
The industry was stuck on a binary: phones on, phones off. Recovery is not binary, it is a gradient. Residents start locked down on day one, earn permissions as they move through program phases, and leave with full access and the habits that come with it. Tiered access was not a feature we added to a phone; it was the mental model that made every other decision fall into place. The first sketch of the permissions matrix took twenty minutes. Convincing the team that this was the real product took longer.
02
Instant Replay, not live monitoring
Staff did not need to watch residents in real-time. That is surveillance, and it creates the adversarial frame the whole category was broken by. What staff needed was a way to triage risk signals after the fact, act on what mattered, and leave residents with dignity in the meantime. Instant Replay became an activity feed staff could scan during rounds instead of a live window into private lives. The design choice kept resident autonomy intact and still gave staff the oversight to intervene early.
03
Replace the clipboard, not the counselor
Recovery center staff already had strong judgment about residents. The portal had to amplify that judgment, not override it. Every permission change, tier promotion, and intervention stayed the counselor’s call. The portal just made those calls faster to make, easier to log, and legible to the rest of the team. I wrote down the decisions the portal would surface, then sat with counselors while they made real calls and watched which signals they actually used. Half the signals I had designed were ignored. The other half became the product.

The deep dive

04 Context

Context

Recovery centers across the country had three bad choices for resident phones. They could ban phones outright, which left residents isolated from family, jobs, and school and pushed them to sneak in hidden devices. They could allow phones completely, which opened the door to old contacts, triggers, and relapse. Or they could repurpose parental-control tools built for teenagers, which were easy for adults to bypass and had no model for recovery progression.

None of these options supported the actual shape of recovery, which is a gradient from high-risk to self-sufficient, not a binary switch. Staff spent hours enforcing no-phone policies instead of doing clinical work. Residents felt surveilled instead of supported. And programs watched early exits climb because residents refused to give up their phones at intake.

Aqua X was built to replace all three of those options with one purpose-built system. A phone for the resident, a portal for the staff, and a tiered access model that adapts as residents progress.

The problemDesign a recovery-specific phone and oversight platform that gives residents structured independence as they move through their program, and gives staff accountability without confiscation.

Where it hurt

Residents Feel IsolatedWhen phones were banned, residents lost contact with family, jobs, and school. The isolation led to frustration, disengagement, and early exits from programs.
Staff Become Phone PoliceEnforcing no-phone policies consumed staff time and pulled them away from treatment and support. The adversarial frame damaged the relationship the program was trying to build.
Existing Tools Were Easy to OutsmartResidents borrowed devices, swapped SIMs, created burner accounts, or reset phones. Parental-control tools built for teenagers were bypassed routinely by adults with more motivation and fewer restrictions on their autonomy.
No Model for ProgressionEvery existing tool locked everything down or left it wide open. There was no way to gradually grant access as residents moved through program phases, so staff had no mechanism to reward progress with more independence.

05 Research

Research

Recovery Center Staff Workflow Mapping

Staff were spending meaningful time every day chasing residents for check-ins, enforcing no-phone rules, and responding to policy violations after the fact.
The "ban phones" default was widely understood to be broken, but no operator had a viable alternative that still preserved accountability.
Counselors trusted their own judgment about residents but had no tooling to support that judgment in real time. Decisions about phone privileges happened on paper, by memory, or in end-of-day meetings.

Partnered on qualitative interviews with directors, counselors, and program staff at partner recovery centersShadowed daily operations to map existing phone policies, check-in workflows, and escalation pathsReviewed incident reports and program-exit data to understand where phone policies failed

Resident Mental Models and Pain Points

Residents did not want unrestricted phones. They wanted earned independence, visible progress, and a sense that the program saw them as adults on a path, not teenagers being managed.
Losing the phone at intake felt like losing identity, continuity with family, and access to jobs or school. Many residents ranked this as the single hardest part of entering a program.
The language of "permission" landed badly. The language of "phase" or "progression" landed well. Residents wanted to know what phase they were in and what phase came next, and what unlocked when they got there.

Partnered on mental-model interviews with residents across multiple centers, focused on phone access, reentry expectations, and program progressionSurfaced the specific moments where residents felt the existing phone policy was unfair, counterproductive, or humiliating

Category Analysis

Parental control tools were built for a different user, a different threat model, and a different caretaker relationship. None of the assumptions transferred cleanly.
MDM platforms gave too much configuration and too little opinion. They required IT expertise the centers did not have.
Nothing in the category had a progression model. Every tool was static: the same rules for day 1 and day 90, with no way to earn more access through program milestones.

Studied existing solutions used by partner centers: consumer parental controls, MDM platforms, and custom rulesetsMapped each solution against the real operational needs surfaced in the staff interviews

06 Design pillars

Design pillars

01Progression Over PermissionAccess adapts as the resident moves through program phases, not as a static setting. The product mirrors the gradient of recovery instead of treating access as binary.
02Accountability Without ConfiscationStaff see what they need to act on without taking phones away. Oversight stops being a punishment and becomes a background condition for trust.
03Counselors Keep the CallEvery permission change, tier promotion, and intervention stays the counselor’s judgment. The portal makes those decisions faster and more legible, not more automated.
04Trust-First, Lockdown as Last ResortThe happy path is designed for dignity. Safety features like nude-photo prevention exist as a last line of defense, not as a surveillance feature surfaced to residents every day.

07 Features

Features

Tiered Phone AccessResidents receive a phone on day one with a tier appropriate to their program phase. Staff assign permissions per tier across contacts, app usage, internet access, location tracking, and camera settings, then promote residents through tiers as they progress. No more all-or-nothing. No more taking phones away entirely.
Instant ReplayFull transparency over messages, searches, and app activity, even after residents erase them. Staff scan an activity feed during rounds instead of watching live, which keeps resident dignity intact while giving early warning on risk signals. Residents keep their phones. Staff still see what matters.
Location TrackingReal-time location visibility so staff never need to rely on residents self-reporting check-ins. Prevents missed check-ins, risky environments, and sneaky meetups. Frees counselor time to do actual counseling instead of chasing accountability.
Nude Photo PreventionComing in 2026. Phones detect explicit content at capture time and lock down immediately. Staff are notified the moment a risky attempt occurs. Prevents exploitation and inappropriate digital behavior without surveilling every photo residents take.
Aqua X Smartphone (Google Pixel 9)A preconfigured flagship device so centers never deal with personal phones, hidden contraband, or IT overhead. Unlimited talk, text, and data on approved contacts. Lifetime device protection built in. If it breaks, Cyber Dive repairs it. If it is lost, Cyber Dive replaces it.
Staff Oversight PortalA web dashboard for counselors and program directors, not for IT admins. Unlimited seat access so every role stays informed. Seamless admin control for real-time permission changes, tier promotions, and intervention logs. On-site setup and staff training included with every deployment.

08 In their words

In their words

Providing a phone on day one, tailored to our program and equipped with boundaries that adapt as residents progress, is a game-changer.
IreneWhere Hope Lives, AZ
Before this, we had no way of knowing what was really happening. Now, we can step in before things spiral out of control.
LandonBob's House of Hope, TX
Before Aqua X, we were constantly chasing down residents who forgot to check in. Now, we know exactly where they are, no excuses, no stress.
HannahZoe Ministries, DE
Nearly all residents in our programs first connect with a predator through their phone. Aqua X eliminates that risk, so healing can come first.
JasminLiving Hope Farms, NH
Aqua X gives residents a way to stay connected without falling back into dangerous patterns. It is not about control. It is about setting them up for success after they leave.
BrianPhoenix Dream Center, AZ
I was reluctant to introduce technology into our program, but once I saw how Aqua X aligns with our mission, I realized it was exactly what we needed for long-term recovery.
YolandaZoe Ministries, DE

Jasmine GraceSafe House ProjectPhoenix Dream CenterZoe MinistriesWhere Hope LivesBob's House of HopeLiving Hope Farms

09 Gallery

Gallery

10 Reflections

Reflections

Alignment is faster than negotiation
At Cyber Dive, redesigning the alert triage workflow, I had three stakeholder groups with conflicting asks. The PM wanted new functionality. Engineering wanted to reduce tech debt in the existing data layer. The founders wanted to ship in six weeks. Instead of negotiating each side separately, I ran a one-hour working session with all three in the room. I walked through the user problem first, then asked the engineer to walk through what data we already had versus what we would need to build. That reframed the conversation. We realized we could surface priority signals inside the existing workflow instead of building a new dashboard. That cut engineering lift in half and still hit the user problem. The shipped version was actually simpler than my original design, and the impact was stronger.
My job is the decision, not the Figma file
I have learned that my job is not to defend my Figma file. It is to get the team to the best decision together, which usually means killing my own ideas when engineering brings better constraints to the table. The tiered-access model looked elegant in my first sketch. The version that shipped was the third sketch, and it came out of a conversation with an AOSP engineer who knew what the launcher could actually enforce at the system level.
The category was wrong before the product was
The biggest design decision on Aqua X was not a screen or a flow. It was rejecting the binary the category had already settled on. Every competing tool was answering the question "phones on or phones off." The right question was "what phase is this resident in." Once that reframe was in place, the rest of the product followed without much argument. Category-level thinking before feature-level thinking is the pattern I keep reaching for.

11 Next in the file

Next projectASU · Mayo Clinic2025ChemoBuddyDesigning for trust in cancer careOpen the case study

All nine case studies