Hey, I'm Riken
Senior AI & Mobile Engineer. I ship production AI products in 4–8 weeks — with the reliability my enterprise background (Ford Mobility, Intuitive Surgical, BCBSM) brings.
Riken Patel
My Story
I'm a Senior AI & Mobile Engineer focused on building real-world AI products with LLMs, with a strong foundation in mobile and product engineering across iOS and Android. Over the past 10+ years I've shipped production software across web, mobile, and AI systems.
My focus is not demos or prototypes. I build systems that hold up in production — low latency, controlled cost, strong privacy, and code that scales with the product. I specialize in shipping AI products end-to-end: streaming LLM pipelines, real-time voice systems, and full-stack applications integrated into real user workflows.
Right now I'm building Intervoo — a real-time AI interview copilot that streams personalized suggestions during live interviews, designed end-to-end with a privacy-first architecture and zero audio storage — and Jishu Labs, a suite of AI-powered developer tools built on a shared backend and design system.
Previously: production systems at scale across mobility (Spin / Ford Mobility), healthcare (Intuitive Surgical — da Vinci robotic surgery), fintech (First Tech FCU), and enterprise platforms (Blue Cross Blue Shield of Michigan). Open to AI product work, consulting, and senior/principal roles focused on shipping real AI systems.
What I Love Building
- Streaming LLM products (OpenAI, Claude) with sub-second latency
- Real-time voice AI with Deepgram and Whisper
- Cross-platform mobile apps with React Native at scale
- Full-stack Next.js with Supabase, Stripe, and AI baked in
- Developer tools that make my own work faster
Education
M.S. in Computer Science
Northwestern Polytechnic University
2014 - 2016 | Fremont, CA
B.Tech in Computer Science
Gujarat University
2009 - 2013 | Gujarat, India
Ten years of shipping taught me a few things I'd stake a project on.
The gap between a demo and a product is where most AI ideas die.
I’ve watched brilliant prototypes fail against real users more times than I can count. The useful engineering work isn’t the first 80% — it’s the last 20% where latency, cost, eval coverage, and failure modes decide whether the thing survives a Tuesday.
Constraints early are a gift. Flexibility early is a trap.
Privacy requirements, a strict latency budget, a fixed scope — the projects where those were defined on day one shipped faster and cleaner than the ones where everything was negotiable. I look for constraints now, instead of trying to design around them.
Premature abstraction costs more than writing the code twice.
Frameworks, orchestration layers, and “future-proof” architectures tend to bill you for complexity you’re not using. I’d rather ship the obvious version, see what actually matters, and factor later. The second version is always sharper than the one written in the dark.
Enforce hard rules at the deepest layer they apply.
Tenant isolation belongs at the database, not at the API. Accessibility belongs at the component primitive, not at QA. Security belongs in the schema of what data exists at all. One layer of real enforcement beats three layers of polite reminders.
I once spent six weeks on infrastructure before shipping anything users could touch.
It was an early solo project. I was convinced I needed a “proper” backend — queues, background workers, a full eval harness, a caching layer — before the first real feature was usable. By the time I shipped, the product hypothesis had shifted and half the infrastructure didn't apply anymore.
Since then I've been ruthless about the rule: build the smallest thing that lets a real user hit the real problem, then let the architecture earn its complexity. On every AI project since, that's been worth more than any clever stack choice.
The stuff that keeps me sharp.
- Trail running and hiking
Fresno puts me an hour from Sierra trails. The best debugging happens somewhere between mile three and mile six.
- Good coffee, serious kitchen
Pour-over obsessive. I like processes with clear variables and immediate feedback — turns out cooking and engineering rhyme a lot.
- Strength training
Four mornings a week. Nothing makes a stuck technical problem feel smaller than an hour under a barbell.
- Reading — slow, annotated, repeat-offenders
Currently rotating through “The Pragmatic Engineer,” anything Will Larson writes, and re-reading “Designing Data-Intensive Applications” for probably the fourth time.
- Family, and long drives with loud music
The non-negotiable one. Everything else is arranged around it.
My Tech Stack
10+ years of collecting tools and shipping production code. AI-first now — here's what I work with daily.
AI / LLM
Tech I Love
Where I've Worked
10+ years shipping software, from healthcare and banking to surgical robotics, e-mobility, and now live AI products.
Founder & Principal Engineer
Built a real-time AI interview copilot end-to-end. Streaming LLM pipelines with sub-second time-to-first-token, real-time voice with browser-side capture and live transcription, and a privacy-first architecture with zero audio storage.
Principal AI Engineer (Contract)
Multi-product AI platform with shared backend, design system, and authentication. Reusable LLM integration patterns (streaming, structured outputs, prompt versioning) and a real-time voice pipeline reused across multiple AI products.
Senior Software Engineer
Delivered rider-facing features for production iOS and Android apps using React Native at large scale. Map-based experiences with Google Maps, deep linking across auth and trip lifecycle, native modules in Swift and Kotlin, CI/CD on Bitrise and GitHub Actions.
Mobile Application Developer
Built secure mobile and web apps supporting surgical workflows for the da Vinci robotic surgery ecosystem. Offline-first data flows, OKTA + encrypted storage, AWS serverless backend with Lambda and API Gateway.
Mobile Application Developer (React Native)
Delivered mobile banking features including payments, transfers, and Zelle P2P. Redux state architecture, Apple/Google Maps integration, third-party SDKs via native bridges, unit and snapshot tests with Jest.
Web Developer (Ionic, IBM MobileFirst)
Built hybrid mobile applications using IBM MobileFirst platform for healthcare services in Michigan. AngularJS + Ionic UI, biometric authentication via Cordova plugins, MVC architecture across components.
Education
Seen enough? Let's build.
10+ years of shipping real software is a good starting point. 30 minutes on a call is usually all it takes to figure out whether we're a fit.