Piyush S.

Most AI projects stall in the same spot. The demo looks great, then someone asks about real users, real data and real rules, and everything grinds to a halt. That gap is where I work. 14+ years of full stack development, 110+ Upwork contracts, 100% Job Success and Top Rated Plus, and the last few years building AI products that live in production: a Medicaid form platform patients finish by chat or voice, a behavioral health EMR with governed AI assistants, healthcare RPA that runs insurance eligibility checks while your staff sleep, and a remote psychological assessment platform. WHAT I BUILD FOR CLIENTS - Generative AI SaaS products and MVPs, from idea to paying users - AI agents, chatbots and voice assistants that answer from your own data (RAG) - Healthcare AI: EMR and EHR integrations, patient portals, FHIR and HL7, HIPAA conscious logging and audit trails - Workflow automation and RPA that removes the copy and paste work - Full stack web apps: React, Next.js, Node.js, TypeScript, Python, FastAPI, PostgreSQL, OpenAI and Claude APIs, Playwright, Temporal, AWS A FEW THINGS I HAVE SHIPPED - AI Medicaid Form Completion Platform: patients answer questions by text or voice, the system prefills from the EMR, generates the government PDF and routes it for approval. - Healthcare RPA and Document Automation: eligibility checks, authorization extraction and medical PDF processing across Availity, UHC and Medicaid portals, with MFA handled. - AI Behavioral Health EMR: clinician AI assistants, an AI receptionist that calls patients, SMS follow ups and full audit trails. - Remote Psychological Assessment Platform: secure onboarding, camera and microphone checks, gaze calibration, video consults and payments. HOW WE WORK TOGETHER - You get one senior engineer who owns delivery, from architecture to deployment - Weekly demos, so you see the product moving instead of reading status emails - Milestones in plain English, with the risky parts flagged early, not on launch day - Replies within a few hours during US business hours - Security first when PHI or sensitive data is involved - 30 days of support after launch, because launch week always finds something If your product has to make frontend, backend, AI, automation and business rules behave as one system, that is my normal day. Send me a few lines about the problem you are solving. You get back an architecture outline, the risks worth knowing about, and a milestone plan, usually the same day. Not sure where to start? Book a short call and we shape it together before you commit a budget.