Epic Panda22
Available for software engineering roles and technical partnerships

Hossain Ahmmed Taufiq · Software Engineer · Technical Founder

Engineering software that organizations rely on.

Building enterprise platforms, AI products and modern web applications with a focus on scalability, architecture and user experience.

From research to production-grade software.

Enterprise SystemsAI EngineeringBackend ArchitectureTechnical Founder

Operations

Production Control

Live

Latency

38ms

Throughput

12.4k

Error rate

0.04%

Request volume

+18%

architecture.ts
1export async function
2 ship(spec) {
3 const arch = await design(spec);
4 return deploy(arch, {
5 mode: "production",
6 observe: true,
7 });
8}

Architecture

Service topology

Full-Stack EngineeringBackend SystemsEnterprise SaaSAI ApplicationsSystem ArchitectureProduction DeliveryTechnical LeadershipResearch to ProductFull-Stack EngineeringBackend SystemsEnterprise SaaSAI ApplicationsSystem ArchitectureProduction DeliveryTechnical LeadershipResearch to Product

About

I build reliable software for real users and real operations.

Full-stack engineer with backend depth, AI experience, and a bias toward clean architecture and shipping.

Software engineer focused on full-stack product engineering, backend systems, and AI applications. I design and ship production platforms — ERPs, CRMs, business dashboards, and AI-powered tools — with an emphasis on architecture, maintainability, and user experience.

I work across discovery, system design, implementation, and delivery. My practice spans enterprise SaaS, research-backed ML systems, and developer-facing products — always oriented toward software that scales and stays operable in production.

Software EngineeringEnterprise SystemsAI ApplicationsSystem ArchitectureProduct EngineeringResearch

Years building

0+

Shipped systems

0+

CGPA

0

Research projects

0

Delivery

CI / CD pipeline

Build
Test
Secure
Deploy
Observe

Role

Software Engineer · Full Stack · Backend · AI

Education

B.Sc. Computer Science & Engineering North South University, Dhaka CGPA 3.83 / 4.00

Availability

Available for senior software engineering, technical consulting, and enterprise / AI builds

Contact

hossainahmmedtaufiq22@gmail.com +880 1728-360834

What I deliver

Enterprise software, AI systems, and modern applications.

The product types I have designed, built, and shipped for clients and teams.

Useful context for roles in product engineering, platform teams, and technical founding work.

01

Enterprise Software

  • ERP
  • CRM
  • CMS
  • MES
  • Business Dashboards

02

AI Systems

  • RAG pipelines
  • Local LLM assistants
  • Automation
  • ML workflows

03

Internal Tools

  • Admin panels
  • Ops dashboards
  • Workflow systems

04

Web Applications

  • SaaS products
  • Marketing platforms
  • Institutional sites

05

Mobile Applications

  • Android clients
  • Cross-channel product surfaces

06

Developer Platforms

  • Desktop tooling
  • Local-first apps
  • API-backed products
Next.jsNode.jsExpressTypeScriptPythonJavaKotlinPostgreSQLMongoDBREST APIsAWSRAG SystemsDockerReactNext.jsNode.jsExpressTypeScriptPythonJavaKotlinPostgreSQLMongoDBREST APIsAWSRAG SystemsDockerReact

Selected work

Production projects with measurable ownership.

Enterprise platforms, AI products, and web systems — links and repositories included for review.

Featured projects

Enterprise Systems

Mango Electric Vehicle Website

Full-stack marketing and operations platform for Mango's electric vehicle brand. Built vehicle showcase pages, reservation and inquiry workflows, and backend APIs on Next.js, TypeScript, Node.js, Express.js, and PostgreSQL — with payment gateway integration on AWS. Focused on Framer Motion-driven UX, SEO, accessibility, and performance optimizations for fast page loads.

Problem

An emerging EV brand needed a production web platform for vehicle discovery, reservations, and inquiries — without sacrificing performance or SEO.

Solution

Shipped a full-stack marketing and operations site with showcase pages, reservation/inquiry workflows, and payment-ready backend services.

Architecture

Next.js + TypeScript frontend, Node.js/Express APIs, PostgreSQL persistence, AWS deployment, Framer Motion for restrained product motion.

Outcome

A public production site supporting vehicle discovery and conversion flows with SEO and accessibility as first-class constraints.

Next.jsTypeScriptNode.jsTailwind CSSPostgreSQLExpress.jsPayment GatewayAWSFramer Motion

Enterprise Systems

SHORBORNO — School ERP Website

Official website for SHORBORNO, a cloud-based School ERP platform serving 100+ educational institutions across Bangladesh. Built with Next.js, TypeScript, Node.js, Express.js, and MySQL over RESTful APIs, with Framer Motion for polished UI — showcasing product features, onboarding flows, and institutional outreach for a nationwide SaaS education platform.

Problem

A nationwide School ERP SaaS needed a professional product site that could communicate capability and support institutional onboarding.

Solution

Built the official SHORBORNO website with product storytelling, onboarding-oriented UX, and a typed full-stack foundation.

Architecture

Next.js + TypeScript UI, Express REST APIs, MySQL, Framer Motion for subtle interaction polish.

Outcome

Product-facing presence for a cloud School ERP serving 100+ educational institutions across Bangladesh.

Next.jsExpress.jsMySQLTypeScriptFramer MotionNode.jsREST API

Web Applications

Playpen School Website

Playpen School Website is a platform that represents the school and allows parents to book school services online. It is built with Next.js, TypeScript, Node.js, Tailwind CSS, PostgreSQL, Express.js, Payment Gateway, AWS, and CDN.

Problem

The school needed a digital presence where parents could discover offerings and book services online.

Solution

Delivered a production school platform with service booking, payment integration, and a maintainable TypeScript stack.

Architecture

Next.js frontend, Express APIs, PostgreSQL, payment gateway, AWS + CDN for delivery.

Outcome

Live institutional website with online booking for admissions, payments, and school activities.

Next.jsTypeScriptNode.jsTailwind CSSPostgreSQLExpress.jsPayment GatewayAWSCDN

Web Applications

NSU ACM Chapter Website

Official chapter website achieving sub-second load times via static rendering and aggressive CDN caching. Clean TypeScript architecture for long-term maintainability.

Problem

The ACM chapter needed an official site that stayed fast, maintainable, and easy to update over time.

Solution

Built a statically rendered Next.js site with aggressive CDN caching and a clean TypeScript architecture.

Architecture

Next.js static rendering, TypeScript, Tailwind CSS, CDN-first delivery.

Outcome

Official chapter website with sub-second load characteristics and long-term maintainability.

Next.jsTypeScriptStatic RenderingCDNTailwind CSS

Additional projects

AI Products

AI CRM — Web, Mobile & Telegram Bot

Full-featured CRM with Gemini 2.5 AI automation for intelligent task management. Extended to a native Android app (Kotlin) and Telegram bot enabling voice and chat-based CRM workflows.

Next.jsExpress.jsMongoDBGemini 2.5Kotlin

Developer Tools

FocusFlow — Desktop App

Fully offline desktop productivity system built with Python, PySide6, and Matplotlib. Integrates task management, habit tracking, Pomodoro timers, notes, calendar, analytics dashboards, and automated backup/recovery into one local-first workspace — no cloud dependency required.

PythonPySide6MatplotlibDesktop

Research

Stock Price Prediction

End-to-end Jupyter pipeline forecasting stock movement by merging Yahoo Finance (BSE) data with Indian news sentiment via TextBlob. Compares ARIMA, LSTM, Random Forest, XGBoost, LightGBM, and more — Random Forest achieved the lowest MAE on the test split.

Pythonscikit-learnTensorFlowTextBlobyfinance

AI Products

LUCY — Local AI Assistant

Local-first J.A.R.V.I.S.-style personal AI that runs fully offline on CPU. Combines FastAPI + Ollama (Qwen3), Whisper STT, Kokoro TTS, and a React UI for chat, live voice, agents, and on-device memory — privacy-first ML product engineering with one-click Windows packaging.

PythonFastAPIOllamaWhisperReact

Experience

Production roles with ownership, delivery, and technical leadership.

Founding engineering, SaaS platforms, and enterprise product work — balanced across architecture, implementation, and client outcomes.

Career profile

Full-stack engineer spanning founding work, SaaS delivery, and enterprise consultancy.

Strongest signals for hiring teams: architecture decisions, shipping under constraints, client communication, and systems that stay maintainable after launch.

04

Roles

18

Impact points

Global

Scope

Focus areas

Enterprise platformsFull-stack SaaSBackend systemsTechnical leadershipClient deliveryAI product engineering

Engagement types

  • Founding & technical leadership
  • Enterprise SaaS product delivery
  • Remote consultancy for US / global orgs
Role 01Bangladesh

Co-Founder & Software Engineer

Softlligence Technologies

Software company focused on enterprise software and AI solutions — leading architecture, product engineering, and client delivery.

System ArchitectureFull-StackEnterprise SaaSAI SolutionsClient Delivery
  • Co-founded a software company focused on enterprise software and AI solutions
  • Lead system architecture and technical strategy
  • Design scalable software systems for ERP, CRM, CMS, and custom business platforms
  • Manage client communication and technical discovery
  • Lead product planning and software engineering decisions
  • Develop production-ready web applications and backend services
  • Build AI-powered automation and business intelligence solutions
Role 02Bangladesh · Software firm

Full-Stack Developer

ZERODEVS LTD

Bangladeshi software firm building SaaS products — developed client platforms including SHORBORNO School ERP and other full-stack web applications.

Next.jsTypeScriptNode.jsExpress.jsMySQL
  • Developed the official website for SHORBORNO, a cloud-based School ERP platform serving 100+ educational institutions across Bangladesh
  • Built full-stack SaaS features with Next.js, TypeScript, Node.js, Express.js, and MySQL over RESTful APIs
  • Delivered product-facing UI with Framer Motion, focusing on performance, clarity, and institutional onboarding
  • Collaborated with the team on delivery timelines, code quality, and iterative SaaS product improvements
Role 03Remote · US-based consultancy

Full-Stack Developer

Brooksource

Enterprise React consultancy delivering solutions to Fortune 500 clients.

ReactTypeScriptReduxAPI IntegrationPerformance
  • Built production React.js + TypeScript applications with Redux for high-traffic enterprise clients
  • Reduced front-end load time by 20% through code splitting, lazy loading, and bundle optimisation
  • Designed API integration layers that measurably reduced cross-service latency
  • Consistently shipped features ahead of schedule across distributed Agile teams
Role 04Remote · Global non-profit

Web Developer

Americares

Global health organisation operating in 25+ countries — mobile-first React applications for staff worldwide.

ReactJWT AuthAPI IntegrationCachingMobile-first
  • Developed mobile-first React.js apps integrating third-party APIs and internal services
  • Reduced backend request failure rate by 25% via error handling, retry logic, and response caching
  • Hardened auth with JWT, input validation, and HTTPS enforcement

Technical skills

Full-stack engineer across web, backend, and mobile.

Production experience with Node.js, Express, Python, Java, Kotlin, and the supporting stack needed to design, build, and ship software end to end.

Categories

8

Technologies

60

Identity

Full-stack

Core stack

What I use most when building production systems

Full-Stack WebJavaScript / TypeScriptNode.js + ExpressPythonJavaKotlin / Android

Active domain

Languages

Core languages across web, backend, and mobile

Browse languages, backend, mobile, cloud, and AI01 / 08

Category

Languages

Polyglot foundation across web, backend services, scripting, and mobile clients.

08

skills

Recruiter takeaway

Full-stack web developer who also ships with Java, Python, Node/Express, and Kotlin — comfortable owning UI, APIs, data, mobile clients, and AI-assisted features.

How I work

Engineering standards that protect quality at scale.

How I approach architecture, delivery, and long-term maintainability on production teams.

01

Clean Architecture

Clear boundaries between UI, domain, and infrastructure so systems stay understandable as they grow.

02

Maintainability

Code should be readable by the next engineer — including future me — without archaeological digging.

03

Scalability

Design for realistic load, graceful failure modes, and growth paths instead of premature complexity.

04

Developer Experience

Good tooling, typed contracts, and predictable workflows reduce friction and raise delivery quality.

05

Performance

Fast interfaces and efficient backends are product features — measured, not guessed.

06

Accessibility

Interfaces should work for more people by default: keyboard, contrast, semantics, and reduced motion.

07

Security

Auth, validation, least privilege, and safe defaults belong in the architecture — not as afterthoughts.

08

Long-term thinking

Ship value now without trapping the system in decisions that cannot be evolved later.

Delivery process

How I take software from discovery to production.

A practical workflow recruiters can map to senior product and platform engineering roles.

Step 01

Discovery

Clarify goals, constraints, users, and success criteria before committing architecture.

Step 02

Architecture

Define service boundaries, data flow, trust zones, and the simplest design that can grow.

Step 03

Database

Model entities, indexes, and consistency needs around real query patterns.

Step 04

Backend

Build APIs, auth, validation, and domain logic with observability in mind.

Step 05

Frontend

Ship clear interfaces with strong hierarchy, accessibility, and predictable state.

Step 06

Testing

Cover critical paths, regressions, and contracts that protect production confidence.

Step 07

Deployment

Automate releases with CI/CD, environment parity, and safe rollout habits.

Step 08

Monitoring

Watch latency, errors, and product signals so issues surface before users do.

Step 09

Maintenance

Iterate with debt awareness — improve operability while continuing to deliver.

Research

Applied research with publication and submission tracks.

Deep learning, multimodal RAG, materials ML, and forecasting — method pipelines and techniques from active and completed work.

Preparing for journal submissionClinical Deep Learning

Epilepsy Detection via Deep Learning on EEG Signals

North South University

Supervisor · Dr. MD. Sumon Hossain

Preprocessing raw EEG signals, designing CNN/LSTM architectures, and evaluating models for automated seizure detection — targeting clinical-grade accuracy benchmarks.

Techniques

CNNLSTMEEG PreprocessingSignal Classification

Method pipeline

1
Raw EEG
2
Preprocess
3
CNN / LSTM
4
Seizure Detect
Preparing for conference submissionMultimodal NLP

Multimodal RAG with Cross-Modal Hallucination Reduction

North South University

Supervisor · Dr. Nabil Bin Hannan

Designed a multimodal RAG pipeline combining text, image, and table retrieval with hallucination-mitigation techniques to improve factual consistency across modalities.

Techniques

RAGCross-Modal RetrievalHallucination MitigationQA

Method pipeline

1
Text · Image · Table
2
Retrieve
3
Ground
4
Answer
PublicationMaterials Informatics

Multi-Task ML for Band Gap & Formation Energy Prediction

North South University

Joint classification + regression model for accelerated materials property prediction — outperforming single-task baselines on key benchmark metrics.

Techniques

Multi-Task LearningRegressionClassificationFeature Engineering

Method pipeline

1
Material Features
2
Shared Encoder
3
Task Heads
4
Properties
PublicationForecasting Systems

Validating Stock Market Price Predicting Algorithms

North South University

End-to-end Jupyter pipeline forecasting stock movement by merging Yahoo Finance historical data (BSE index) with Indian news headline sentiment via TextBlob polarity and subjectivity scores. Workflow covers data collection, EDA, stationarity testing, date-based feature merging, and model comparison — ARIMA baseline, LSTM multi-step forecasting, and regression models including Random Forest, XGBoost, LightGBM, AdaBoost, and Decision Tree. Random Forest achieved the lowest MAE on the test split among evaluated models.

Techniques

ARIMALSTMRandom ForestSentiment Analysis

Method pipeline

1
Market + News
2
Features
3
Model Suite
4
Forecast

Education

Computer Science foundation with strong academic performance.

Coursework across algorithms, systems, software engineering, and machine learning.

Bachelor of Science in Computer Science and Engineering

North South University

  • CGPA: 3.83 / 4.00
  • Relevant coursework: Data Structures, Algorithms, Software Engineering, Database Systems, Machine Learning, Artificial Intelligence, Operating Systems, Computer Networks
  • Research: Epilepsy Detection (Deep Learning) · Multimodal RAG (NLP/CV)

HSC, Science

Notre Dame College, Dhaka

  • GPA: 5.00 / 5.00
  • Volunteered in organizing and hosting college events
  • Developed teamwork, leadership, public speaking, and event management skills

Certifications

  • Full-Stack Web Dev (Programming Hero)
  • DSA (Udemy & Phitron)
  • AI & ML (GP Academy, Programming Hero, Udemy)
  • Public Speaking (BYLC)
  • Bash Scripting (freeCodeCamp)
  • Arduino (Gobeshona)

Languages

  • BanglaNative
  • EnglishProfessional fluency (written & spoken)

Highlights

Evidence across engineering, leadership, research, and academics.

Concrete signals for hiring conversations — no inflated claims.

Engineering

03 items
  • Shipped production platforms for education, EV, and SaaS clients
  • Built multi-channel AI CRM spanning web, Android, and Telegram
  • Designed local-first AI assistant with CPU-optimized inference packaging

Leadership

03 items
  • Co-founded Softlligence Technologies
  • Led architecture and technical strategy for enterprise/AI product work
  • Owned client discovery and product planning conversations

Research

03 items
  • Active deep-learning research on EEG-based epilepsy detection
  • Multimodal RAG with cross-modal hallucination reduction
  • Published multi-task ML and stock prediction research work

Academic

03 items
  • B.Sc. CSE at North South University — CGPA 3.83 / 4.00
  • HSC Science — GPA 5.00 / 5.00 (Notre Dame College)
  • Relevant depth across systems, ML, and software engineering coursework

Contact

Open to roles where strong engineering ownership matters.

Available for full-stack, backend, AI, and technical consulting opportunities. Happy to discuss product scope, architecture, and delivery timelines.

Senior Software EngineeringTechnical ConsultingEnterprise SoftwareAI SystemsStartup MVPsFull Stack Development