AI-powered enterprise platform

We build intelligent systems for business growth

DewanAI is a custom AI development studio. We partner with businesses to design, build, and deploy AI-powered applications — from LLM document intelligence to live market data platforms and automated analytics workflows.

Live & beta tools
14
Working software, not mock-ups
Data sources
17
Monitored in real time
Economic events
60+
BLS, BEA, and Fed releases
Companies covered
40+
Across 8 market sectors

What we've built

Platform capabilities

Every module below is working, deployed software. Each one solves a real business problem with AI and data.

RAG Document Intelligence

Beta

Upload PDFs, Word docs, or Excel files and ask natural-language questions. The RAG pipeline retrieves the most relevant passages and returns cited answers — no hallucination, no guesswork.

LLMVector SearchPDFEnterprise

AI Conversational Agent

Beta

Multi-turn enterprise chatbot with tool-use: queries databases, calls REST APIs, reads live market data, and executes multi-step workflows — entirely through plain English.

LLMMulti-AgentNLPAPI Integration

Market Intelligence Dashboard

Live

Live price quotes, % change, sentiment scoring, and AI-narrated outlooks for the S&P 500, Nasdaq, Mag-7, healthcare, and energy sectors. 4-hour intelligent cache.

Yahoo FinanceSentimentFintechReal-time

Market Calendar

Live

Unified economic + earnings calendar. FOMC decisions, CPI & NFP releases, GDP estimates, options expiration, and triple witching — all sourced from BLS, BEA, and the Federal Reserve.

BLSFederal ReserveBEAyfinance

Earnings Intelligence

Live

Track earnings performance for 35+ public companies. View quarterly revenue, EPS, net margin trends, YoY comparisons, composite performance scores, and automated data refresh from Yahoo Finance.

Yahoo FinanceEarningsFintechAnalytics

Business Intelligence Suite

Live

Full-cycle business analytics platform. Track revenue, margin, expenses, and customer KPIs. Compare against industry benchmarks, run AI-powered pipeline ingestion, and surface executive insights.

KPIAnalyticsPipelineBenchmarking

Investment Research Portal

Live

Coverage of 40+ public companies across 8 sectors. Live price data, YoY growth, investment signals (Bullish/Bearish/Caution), cash flow indicators, and AI-powered sector summaries.

Equity ResearchSectorsSignalsyfinance

Weather Intelligence Dashboard

Live

Live NYC weather powered by the Open-Meteo API with 7-day forecasts, historical 50-year records, AI event summaries, comfort scoring, and real-time conditions monitoring.

Open-MeteoForecastHistorical DataAI Insights

Inventory & Operations

Live

End-to-end inventory management: products, stock levels, purchase orders, vendors, and transaction history. Real-time stock alerts, low-inventory warnings, and procurement tracking.

InventoryPO ManagementVendorsOperations

Pipeline Monitor

Live

Central observability dashboard for every DewanAI data source. Real-time online/offline status, latency tracking, cache freshness, scheduled data updates, and full API documentation.

MonitoringObservabilityDevOpsAPIs

Prospects & CRM

Live

Lightweight CRM for managing leads, clients, and outreach. Prospect intake forms, pipeline tracking, email template engine, campaign logs, and one-click email delivery via SendGrid.

CRMEmailSendGridOutreach

Document Analysis (GPT Pipeline)

Live

Runs documents through a multi-step ChatGPT analysis pipeline. Extracts key topics, summaries, action items, and structured insights from unstructured text.

GPT-4NLPDocument AI

AI User Stories Generator

Live

Converts raw feature ideas or requirement notes into properly written Agile user stories with acceptance criteria — saving hours of BA and product-owner time.

GPTAgileSDLC Automation

AI Prompt Studio

Live

Interactive prompt engineering workbench. Design, test, and compare system prompts, temperature settings, and model parameters — ideal for fine-tuning GPT-powered workflows.

Prompt EngineeringGPTDev Tool

AI HR & Recruitment Intelligence

Soon

Parse resumes, score candidates against JDs using embeddings, rank shortlists, generate skill-gap reports, and draft interview questions — dramatically reducing recruiter screen time.

NLPEmbeddingsHR TechRanking
In development

Under the hood

Technology stack

A modern, production-grade stack chosen for reliability, speed, and minimal operational overhead.

  • Zero vendor lock-in — open-source core
  • Swap any layer (SQLite → PostgreSQL, Flask → FastAPI)
  • Auth, caching, scheduling, and API layers already in place
  • Blueprint architecture — add modules without a refactor

AI & Language Models

OpenAI GPT-4 / GPT-4o
LLM backbone for analysis, generation, and agent reasoning
RAG / Vector Search
Retrieval-Augmented Generation for grounded, cited answers
Embeddings & Cosine Similarity
Semantic search and candidate matching
Prompt Engineering
Structured prompting, chain-of-thought, and tool-use patterns

Backend & API

Python 3.14
Core application language
Flask 3.x
Lightweight web framework with Blueprint architecture
Flask-Login & Flask-SQLAlchemy
Authentication and ORM
Flask-Mail / SendGrid
Transactional email delivery
ThreadPoolExecutor
Parallel data fetching for low-latency pages

Data & Finance APIs

yfinance 1.2
Yahoo Finance — live prices, earnings dates, news headlines
BLS (Bureau of Labor Statistics)
Official CPI, NFP, PPI, JOLTS release dates
BEA (Bureau of Economic Analysis)
GDP advance and revised estimates
Federal Reserve
FOMC meeting schedule and rate decisions
pandas
Data wrangling, time-series manipulation, and aggregation

Frontend & UI

Bootstrap 5.3.3
Responsive grid, components, and utility classes
Bootstrap Icons 1.11.3
1,800+ SVG icons — zero external image dependencies
Jinja2 Templating
Server-side template inheritance and dynamic rendering
Vanilla JavaScript (ES6+)
Tab navigation, live refresh, filter interactions
DewanAI design system (v3)
App shell with sidebar, ⌘K command palette, toasts, and shared UI macros

Infrastructure & Storage

nginx → gunicorn → Flask
TLS-terminating reverse proxy in front of multi-worker WSGI on AWS EC2
SQLite (WAL) or PostgreSQL
SQLAlchemy ORM — switch databases with a single DATABASE_URL setting
Shared file cache
Warm market, weather, and calendar data shared by every worker (instance/cache)
Scheduled data refresh
flask update-data --due on a systemd timer keeps every data set fresh
RESTful JSON APIs
Internal /api/* endpoints consumed by the UI and integrators

Getting started

How we onboard a new client

From first conversation to a live production system — no surprises, no black-box hand-offs.

Typical engagement

Discovery to proposal1–3 days
Integration setup3–5 days
First working prototype1–2 sprints
Production deployment4–8 weeks total
Hyper-care support30 days post-launch
  1. 01

    Discovery Call

    We start with a focused 30–60 minute call to understand your business challenge, data sources, team size, and success metrics. No NDA required at this stage.

    Shared problem statement & goal definition
  2. 02

    Demo & Fit Assessment

    We walk through the live DewanAI demos most relevant to your use case. You see working software — not slides. We identify which modules to customize or build fresh.

    Recommended solution architecture
  3. 03

    Scope & Proposal

    We deliver a written proposal covering: solution scope, deliverables, integration points, tech stack, timeline (sprint-by-sprint), and a fixed or T&M pricing model.

    Proposal document + project timeline
  4. 04

    Access & Integration Setup

    We provision the client workspace, connect to your data sources (APIs, databases, file stores), configure SSO/LDAP if required, and establish a dev → staging → prod pipeline.

    Working dev environment with real data
  5. 05

    Build & Iterate

    2-week agile sprints with mid-sprint checkpoints. You see progress in working demos every week. Feedback loops are tight — pivots are cheap early, not after go-live.

    Tested, reviewed feature increments
  6. 06

    Deploy, Train & Hand Off

    Production deployment to your cloud or on-prem environment. We provide user training, admin documentation, runbooks, and a 30-day hyper-care support window.

    Live system + knowledge transfer

Our approach

Built different

Code-first, always

We ship real, working software — no low-code black boxes or vendor-locked platforms. You own every line we write.

Fast iterations

Two-week sprints with working software every cycle. Decisions are made on running demos, not architecture diagrams.

Security by design

Auth-gated routes, hashed passwords, input validation, and parameterized queries are defaults — not add-ons.

Contact the founder

Get in touch

Scan the code or reach out directly — it opens this page on your phone.

DewanAI
AI Solutions Studio

Mamun Hossain

Founder & AI Solutions Architect

“We build intelligent systems for business growth”

Mamun0715@gmail.com 929-216-8766
Bay Shore, NY
QR code linking to https://dai.reservedrive.com/about
Scan to open
DewanAI
https://dai.reservedrive.com/about
Start a conversation

Ready to build something intelligent?

Tell us your challenge. We'll identify the right AI solution, show you a live demo, and have a proposal to you within 48 hours.