.png&w=256&q=75)
ROBIUL HASAN@rhasan22959110
8
Events attended
2
Submissions made
Bangladesh
1 year of experience
About me
I’m Robiul Hasan, a support engineer and builder focused on practical AI-powered products, SaaS platforms, and business automation. I combine hands-on IT infrastructure experience with a strong interest in turning complex workflows into simple, useful software. My background includes Microsoft 365 and Azure administration, technical support operations, cloud-based systems, and full-stack product development using modern tools such as Next.js, Supabase, PostgreSQL, and AI coding platforms. I’m currently building products for small businesses, including finance, inventory, ordering, and operational-management solutions. At hackathons, I enjoy working on ideas that solve real operational problems: AI assistants, workflow automation, intelligent dashboards, document processing, and tools that make technology more accessible for teams and customers. I value clear product thinking, rapid prototyping, reliable implementation, and solutions that can move beyond a demo into a usable product.
🤓 Latest Submissions
MediScribe Live
Emergency clinicians lose a large part of every shift to typing EHR notes. AI scribes can help, but a clinician cannot trust a note if they cannot see where each line came from. MediScribe Live has two voice modes that share one patient chart: Bedside Voice Triage (AssemblyAI Voice Agent API). A full-duplex voice agent talks with the patient, handles barge-in, and calls tools while it talks: record_patient_intake, check_drug_interaction and flag_critical_vital. The chart, the ESI acuity score and the alerts update live. The panel shows the real response latency, measured in the browser from the end-of-speech event to the first agent audio. Ambient Scribe (AssemblyAI Realtime STT). Universal-3.5 Pro with domain: medical-v1 and speaker labels transcribes the consultation and separates Doctor and Patient. Source-linked SOAP note (AssemblyAI LLM Gateway). The server numbers every transcript turn. Each SOAP item cites the turns it came from, and the server checks each citation against the real turn text. Wrong citations are fixed or dropped. Items with no source are flagged "NO SOURCE - VERIFY". Nothing is invented: topics not discussed say "Not discussed during encounter", allergies never default to NKDA, and a diagnosis appears only if the clinician said it. Clinician sign-off and FHIR. The note stays a DRAFT until a named clinician signs and acknowledges any unsourced items. Export is an HL7 FHIR R4 Bundle whose Composition moves from preliminary to final with an attester. Security: the API key stays on the server. The browser gets short-lived tokens and connects straight to AssemblyAI over WebSockets. Stack: Next.js 14, React, TypeScript, Tailwind, Web Audio API, AssemblyAI Voice Agent API, Streaming STT v3, LLM Gateway. Deployed on Vercel. Note: this is a hackathon prototype, not a medical device. The drug-interaction list is a small demo set.
30 Sep 2026
LedgerBuddy: Digitizing Emerging Markets with AI
Inspiration Across Bangladesh and other emerging markets, many SMEs still rely on paper ledgers, printed invoices, and handwritten receipts. Manual entry is slow and error-prone, leaving owners without a clear view of cash flow, inventory, or profit. We already build LedgerBuddy, an offline-first POS and inventory app. LedgerBuddy AI solves one major friction point: turning supplier invoices and receipts into usable business data without manual typing. What it does LedgerBuddy AI is a Bengali-first Smart Ledger and Business Operations Copilot. Users upload an invoice, receipt, or expense bill. The app extracts supplier details, line items, quantities, prices, totals, and payment status into structured ledger records. It validates totals, flags low-confidence fields, duplicate invoices, and mismatches, then lets the user review and approve each entry. Its AI Copilot understands the digitized ledger and responds in Bengali or English. A shop owner can ask, “এই মাসে আমার সবচেয়ে বেশি খরচ কোথায় হয়েছে?” and receive a data-backed answer with metrics, risks, and recommended actions. How we built it The frontend uses React, Vite, and Tailwind CSS v4. Supabase provides PostgreSQL, Storage, and Edge Functions. Fireworks AI powers document interpretation and multilingual reasoning through a resilient model waterfall: it prioritizes a custom-deployed model for Bengali/English invoice parsing, then falls back to DeepSeek-v4-Pro. AI handles extraction and explanation; totals, validation, stock calculations, and financial metrics remain deterministic and auditable. What’s next LedgerBuddy AI will integrate with the LedgerBuddy Flutter POS app and its Drift SQLite sync engine. Shop owners will be able to photograph invoices on Android, create local purchase orders with unreliable connectivity, and ask questions about offline-first business data.
13 Jul 2026
