iOS · Android
Boyfi
AI companion for meaningful conversations and immersive role-play.
Senior Independent Mobile Consultant
Mobile MVPs for startups —
built by a senior developer, without the agency overhead.
I'm Mikalai Siliuk — an independent mobile consultant with 10+ years in iOS, Android, and Flutter. I personally lead architecture, development, and release. When a project needs design, backend, or QA firepower, I bring in trusted specialists from my network.
10+ years · 100+ projects · Flutter · iOS · Android · Replies within 24 hours
Who I help
Three types of teams
get the most out of working with me.
🚀
Startup founders
Turning a validated idea into a production-ready mobile MVP — architecture, App Store / Google Play launch, subscriptions, analytics, growth instrumentation from day one.
🤝
Agencies
A senior mobile contractor you can put in front of your clients. NDA-friendly, predictable delivery, clean handover. White-label or co-branded.
📈
Scale-ups & enterprise
Refactoring, architecture audits, performance work, crash-free release process, and senior engineering capacity for mature mobile products.
Founder testimonials
What founders say —
and what their shipped apps say for them.
Three founder quotes from NDA projects, plus three publicly verifiable testimonials from apps live on the stores right now. Install any of them and judge the work yourself.
We had 8 weeks before our seed investors expected a working demo. Mikalai took our Figma designs and delivered a production-ready Flutter app — subscriptions, analytics, and crash reporting all wired in from week one. Eighteen months later, the codebase is still clean enough that our in-house team maintains it without touching the architecture. I’ve recommended him to three other founders since.
CEO, AI companion app
United States · 2024 · under NDA
I’m not a developer, so I was genuinely scared of picking the wrong person and burning through my runway. Mikalai made every decision visible — weekly builds I could install on my own phone, clear explanations of trade-offs, written notes on every architectural call. We went from a validated idea to live on the App Store in 11 weeks. The process was the least stressful part of building the company.
Co-founder, health & wellness app
United Kingdom · 2025 · under NDA
We needed two mobile products simultaneously: a field inspection tool for engineers in remote locations with offline-first requirements, and an executive dashboard for our board to track directives. Mikalai architected and shipped both — on time, no critical bugs post-launch. The field app handles daily active users across multiple regions; the board dashboard is part of our standard governance process. When a mobile product cannot fail, you call someone like Mikalai.
Head of Digital Transformation, oil & gas company
Europe · 2024 · under NDA
Publicly verifiable
Founders whose apps you can install today.
Boyfi — AI Companion App
Flutter · iOS & Android · OpenAI · Grok · Adapty
We had a Figma file, a tight investor window, and a clear product idea — but the technical scope was unusual: dual subscriptions running in parallel and two AI providers we needed to keep swappable. Mikalai scoped the whole problem in one call, designed the abstractions up front, and shipped on Flutter for iOS and Android. Months later, we added a third AI provider by touching one module. The architecture has kept pace with every product change since.
MyCoach — Fitness Training App
Flutter · iOS & Android · Coach-client architecture
As a non-technical founder, the part I worried about wasn’t the build — it was the decisions I couldn’t evaluate myself and would have to live with for years. Mikalai walked us through the architecture in plain language before writing any code: why a coach-client app needs a specific data model, what breaks if you skip it, what we’d regret in eighteen months. Then he built it exactly that way. Two years and sixty updates later, the structure still holds.
AI Music & Song Generator
Flutter · iOS · FoxAI · MusicAPI · Qonversion
We wanted an iOS app that turns a few lines of text or full lyrics into a complete song — generation behind a paywall, attribution wired in, App Store featuring-ready. Mikalai built the AI provider abstraction first so we could plug FoxAI and MusicAPI in parallel, then shipped the full flow with Cloud Functions handling the generation queue. Subscriptions, analytics, and the whole creative experience went live without a single App Store rejection.
Live on the stores
Every portfolio card links to a public store listing — install, review, judge the work yourself.
Numbers with sources
Metrics in case studies are tied to a source: Stripe, Agora telemetry, Firebase, Amplitude.
A decade of shipping
10+ years in mobile. Native Android since 2015, iOS since 2018, Flutter since 2020.
AI-augmented delivery
Ship 2× faster.
Without the vibe-coder discount.
I use AI tooling everywhere it earns its keep — and stay in the driver's seat for everything that matters: architecture, edge cases, production quality.
2–4 wks
Faster MVPs
−50%
Lower budget
100%
Same quality bar
Repo-aware coding partner
Claude Code
AI that reads your whole codebase before writing — refactors, migrations, and feature scaffolds in minutes instead of days.
AI IDE for tight loops
Cursor / Windsurf
Inline generation and multi-file edits with project context. Where Claude Code does the heavy lifting, Cursor handles fast iteration.
Design-to-code, no copy-paste
Figma MCP
Designs come straight from Figma into Flutter / SwiftUI screens — pixel-faithful, with the project's tokens and components reused.
PR audits before merge
AI code review
Every pull request gets an automated senior-level review — security, regressions, dead paths — before a human reviewer touches it.
01
AI scaffolds, I architect
Boilerplate, models, API clients, basic screens — generated in hours. The architecture, the data model, the tradeoffs — decided by me upfront, not improvised by an LLM.
02
AI iterates, I review
Feature work happens in tight AI-assisted loops. Every commit goes through senior review and AI-powered PR audit. Production-grade tests, not toy demos.
03
AI accelerates, I ship
Release pipelines, store listings, privacy manifests — automated where it's safe, hand-checked where it counts. The last 10% to launch stays human.
How we'll work
From validated idea
to App Store in 2–4 weeks.
A predictable AI-augmented rhythm built for founders who need progress every week. Weekly builds you can install, written decisions on every architectural call. Larger, production-grade scope flexes to 4–6 weeks — same senior quality bar.
01Week 1
Discovery & architecture
Discovery call, architecture proposal, and milestone plan. AI-assisted scoping accelerates the work; the architectural decisions stay with me. You leave the first week with a written scope and a roadmap to the App Store.
02Weeks 2–3
AI-accelerated builds you can test
Production-grade Flutter or native code from day one. Claude Code, Cursor, and Figma MCP compress the implementation; CI/CD, analytics, subscriptions, crash reporting wired in. Weekly TestFlight / internal builds you install on your own phone.
03Week 4
App Store & Google Play launch
Store listings, privacy manifests, review responses, phased rollouts. I own the last 10% to public launch — no surprises, no rejections.
04After launch
Stabilise, hand over, scale
2–4 weeks of post-launch stabilisation — crash monitoring, store reviews, hotfixes. Then either a monthly retainer or a clean handover with architecture docs and a runbook.
Deep dive
One project,
in depth.
A real-time communication platform connecting users with live human companions — mobile app, web operator panel, cloud backend. Shipped and scaled under NDA.
NDAReal-Time Companion
Real-Time Companion
Communication Platform
I led delivery of a three-product platform — mobile app, web operator panel, and cloud backend — that matches users with trained human companions for real conversations over text, audio, and video.
My role · Lead mobile developer · Architecture · Release owner · Under NDA
Read the case study
1,000+
Daily Active Users (Amplitude, 90d rolling)
3,000+
Calls per Month (Agora RTC telemetry)
99.7%
Uptime (Firebase health checks, 90d)
+40%
Revenue growth, 90d after V2 (Stripe)
Curious how this kind of architecture survives years of feature work? See how AI-augmented delivery keeps it clean.
Published
Apps shipped to the stores.
A selection of mobile apps delivered end-to-end — from idea and architecture to publishing and analytics.
iOS · Android
MyCoach
Personalized workout plans, recovery tracking, and 250+ exercises.
iOS
AI Music & Song Generator
Turn prompts and lyrics into complete original songs, instantly.
iOS
RoomFlash
Transform any room photo into a fully redesigned interior with AI.
iOS
Coin ID Scan
AI-powered coin recognition with PCGS grading and live market values.
iOS
Screen Time Limit: Focus Mode
Block apps, set limits, and build focus sessions — with parental controls.
iOS
Mushroom Identifier
Identify mushroom species from a photo with AI and a curated database.
iOS
Plant ID — Flower Care
Identify plants from photos and get personalized care recommendations.
iOS
Hairly — Hair Color Changer
Try 30+ hairstyles and colors instantly with AI-powered virtual try-on.
Why independent, not an agency
This is the question
every serious client asks —
here's the honest answer.
Direct line to the engineer
No account-manager layer, no sales-to-engineering translation loss. When you message me, you're talking to the person writing the code and making the architectural calls.
Senior-level decisions on every call
In agencies, senior engineers set direction and junior engineers ship it. Here, the same person does both. Slower than a 10-person team on paper — but produces fewer costly rewrites.
Agency-grade depth, on demand
When a project needs design systems, heavy backend, or dedicated QA, I plug in trusted partners I've shipped with before. Specialist depth without full-agency overhead or lock-in.
Scaling up isn't a blocker: if a project needs a bigger team — extra mobile engineers, backend, design, or QA — I plug in teams I've already shipped with. You don't have to go hunting for vendors, and I don't have to step off the project. Same point of contact, same delivery standard, just more hands.
Frequently asked
What founders ask before
the first call.
- How long does an MVP take?
- 2–4 weeks for most AI-augmented MVPs, with weekly builds you can install and test throughout. Larger, production-grade scope ships in 4–6 weeks. Same senior quality bar — AI accelerates the work that doesn't need senior judgment.
- What if I need to scale the team later?
- When a project needs a second mobile engineer, backend, or dedicated QA, I plug in trusted partners I've shipped with before. You keep one point of contact and don't go hunting for vendors.
- What happens after launch?
- I stay on for 2–4 weeks of post-launch stabilisation — crash monitoring, store review responses, hotfixes. After that, I either move to a monthly retainer or hand the codebase over with architecture docs and a runbook.
More on engagement, pricing, and NDAs.
Send a message
Have an idea?
Let's make it real.
From a first sketch to a live app on the stores. Reach out through the channel that's easiest for you — I reply within 24 hours.