
OncFlow
Clinical AI platform for oncology workflows.
Built primarily by me with support from one junior engineer—backend, AI workflows, FHIR integration, multi-tenant SaaS, and cloud deployment.

Alok Dwivedi
I enjoy taking ideas from a blank page to working software.
Over the past year I’ve built multiple production-ready AI applications across healthcare, education and creative tools, combining 25 years of distributed systems experience with modern AI engineering.
My background in enterprise architecture helps me build reliable systems—but my passion is building products.
I enjoy:
Things I’ve designed and shipped. Try the demos where they’re live.

Clinical AI platform for oncology workflows.
Built primarily by me with support from one junior engineer—backend, AI workflows, FHIR integration, multi-tenant SaaS, and cloud deployment.


AI-assisted oral cancer screening with privacy-first on-device ML.
Flutter application, ONNX inference, and a CNN ensemble that runs on device.

Adaptive literacy platform that generates curriculum-aligned reading content with AI.
Adaptive engine with Flutter clients, Python backend, and Firebase.

Collaborative AI storytelling platform.
Focused on context engineering, memory, multi-turn reasoning, and story consistency.
A lightweight loop. AI accelerates implementation, but engineering judgement stays essential.
Evidence over resumes.
End-to-end products that real people use.
Shared infrastructure with solid tenancy and ops.
Models wired into product flows, not demos in notebooks.
Multi-step reasoning, tools, memory, and guardrails.
APIs, data, and services built to stay up.
Flutter apps for iOS, Android, and the web.
Tools that make builders faster and clearer.
Clinical and patient-facing systems with care for privacy.

I’ve spent over twenty-five years building distributed systems for companies like Veritas, Maersk, Discover, and OpenText—platforms that had to work under real constraints of scale, performance, and security.
That background still shows up in how I design software: how systems fail, where guardrails belong, and what “production-ready” actually means. Today I apply it to AI-powered products—LLMs in real workflows, multi-stage reasoning, and shipping software people can trust.
What excites me most is taking an idea from a blank page to software that people actually use. I spend most of my time designing systems, writing production code, and iterating with real users to make the product better.
If you’re looking for a builder—not a consultant—I’d like to hear from you.
Email info@3amwave.com