3 Products Live

AI That Builds What Matters.

Purpose-driven intelligence. Built for real-world impact.

Independent AI research and open-source products. A personal lab by Mohammed Jawed — building AI for problems that genuinely matter, across healthcare, security, finance, and more.

Explore Products
mannatai.com

$ initialize_ai_system --all

 HAKIM AI Companion   healthcare

 KavachAI Security    security

 Mizaan Scoring       ● LIVE

 Command Center      sales● LIVE

 Muneem Ji billing    ● LIVE

 Quillra publishing   SaaS

 ArkThor C2 Detect    security

 HAKIM ThreatSense    security

➜ 8 systems initialized

$ _

 3 live  5 in dev  Claude API
8+ AI Products Built
IIT-K+TU Wien Alumni
18+ Years in IT
Why Us

What this lab
is about.

A personal AI lab. Open-source first. Products that ship and stay shipped.

Research Pedigree

IIT Kanpur, India & TU Wien, Austria & A.M.U India academic foundation. Architectural decisions are grounded in current research — not trend-chasing.

Meet the founder

Security by Default

The lab includes a dedicated AI security product (KavachAI). Every system ships with threat modelling, input validation, and full audit trails.

See KavachAI

Ships. Not Slides.

8 live or near-live AI products already. The work in this lab is built end-to-end — code, deployment, and the long tail of keeping it running.

View all products
0 AI Products Built
0 Live in Production
0 Years AI + Security
100% Real-World Focus
Our Products

Applied AI That Just Works

We assemble and adapt cutting-edge AI tools to solve real problems in healthcare, education, cybersecurity, and beyond.

Sales Tech Live

Command Center

Real Estate Sales Execution

Command Center · Dubai Brokerage ● Live
24 Active Leads
→
6 Follow-ups Due
→
3 Closing This Week
Call back: Al Barsha lead — 48h overdue Urgent
Send proposal: Downtown unit 2204 Pending
Viewing confirmed: JBR penthouse Done

The CIA of real estate execution — transforming raw leads into closed deals through Control, Intelligence, and Action. Live and actively used by a Dubai-based real estate brokerage.

CIA3-Pillar
Lead→Gold Pipeline
●Active in Dubai
AI Scoring Live
Mizaan — multi-agent AI scoring engine logo

Mizaan

Multi-Agent AI Scoring Engine

Mizaan · Score Report ● Live
Primary Agent
87%
Auditor Layer
94%
Human Validation
72%
3-Layer Pipeline RAG Learning Explainable

Mizaan (meaning "Balance" in Arabic) is a multi-agent AI scoring system that measures impact — not just volume. A three-layer pipeline (Primary scorer → Auditor → Human validation) delivers transparent, explainable performance evaluation across any domain, with RAG-based continuous learning.

3-LayerAgent Pipeline
RAGContinuous Learning
Plug-inSDK, API available
FinTech · Open Source Live

Muneem Ji

GST Billing for Indian SMBs

Muneem Ji · Invoice #INV-2024-0142 ● GST Filed
Customer NEW HODA ENTERPRISES
Items: 12 Subtotal: ₹4,250
CGST 9% ₹382
SGST 9% ₹382
Total: ₹5,014
Print Share GSTR-1 Export

Self-hosted, MIT-licensed GST billing for Indian small businesses — kirana, pharmacy, electronics, restaurant, jewellery, auto parts. Built for shops that need real software without SaaS subscriptions or vendor lock-in. One Docker command to install. Runs on a mini-PC, a Raspberry Pi, or any small VPS.

MIT Licensed
Self-Hosted
Docker Ready
12+Modules
GSTR-1 / 3BExports
●Live in India
Publishing SaaS
Quillra — multi-tenant blog SaaS logo

Quillra

Multi-Tenant Blog Platform

Quillra · Editorial Workspace ● Phase 1
Draft · "Designing for trust in AI products" In Review
Author: Priya S. · Org: AcmeBlog Pending
Published · "Quarterly product notes" Live
Multi-Tenant Author / Admin Roles Reader Pages

Multi-tenant SaaS where each organization runs its own blog — authors write, admins review and publish, readers consume through clean public pages. Think Medium's editor + a Contentful-style admin layer + a Notion-style UI. Roadmap: AI assistance (Phase 2), public REST API + embeddable widgets (Phase 3), custom domains (Phase 4).

SaaSMulti-Tenant
4-PhaseRoadmap
In DevPhase 1
Healthcare AI

HAKIM AI Companion

Pediatric Healthcare Assistant

HAKIM · Patient Consult
Child · 7 yrs · Fever 38.9°C · 2 days
⚒ HAKIM Likely viral infection. Confidence 94%. Refer if fever persists > 3 days.
Confidence
94%

Revolutionary pediatric AI assistant fusing 5,000 years of Ayurvedic wisdom with cutting-edge diagnostics and multimodal analysis — offline-ready, telemedicine-integrated, globally scalable.

OmniHealth Triage
24/7AI Care
GlobalReach
AI Security
🔬

KavachAI

AI Security Evaluation

KavachAI · Threat Scan
⚠ Prompt injection detected HIGH
✓ Data exfiltration attempt BLOCKED
✓ Jailbreak pattern BLOCKED
Risk Score: 2.1 / 10  ·  Safe

Combines adversarial testing, code-level risk detection, and LLM vulnerability scanning to ensure your AI systems are resilient and production-ready.

3-in-1Tools
0-100Score
∞Plugins
Threat Intel

ArkThor AI

C2 Detection Platform

ArkThor · PCAP Analysis ● Scanning
📄 capture_2024_03.pcap 4.2 MB · 18,340 pkts
⚠ C2 beacon — Cobalt Strike 93% conf.
⚠ DNS tunnelling detected HIGH
✓ Lateral movement — none CLEAN
Threat Score: 8.4 / 10  ·  Critical

Detects, classifies, and analyzes malware threats and C2 traffic by inspecting .pcap files using advanced machine learning algorithms with 93% accuracy.

93%Accuracy
0-DayThreat Shield
AI-C2Traffic Intel
Cybersecurity

HAKIM ThreatSense

AI Cybersecurity Platform

ThreatSense · SOC Monitor ● Live
3 Critical
7 Warnings
142 Resolved
99.8% Uptime
⚠ Brute-force: SSH port 22 ACTIVE
✓ Ransomware signature match BLOCKED
▶ Gen-AI Isolate affected node & rotate SSH keys

Enterprise-grade AI threat detection designed for NGOs and startups — at a price they can afford. Real-time monitoring, intelligent threat analysis, and guided incident response powered by Generative AI.

3-TierAI Defence
Gen-AIIncident Help
GuidedResponse
Python LangChain FastAPI React Docker PostgreSQL OpenAI API Claude API TensorFlow AWS Redis Kubernetes RAG Pipelines Vector DBs Python LangChain FastAPI React Docker PostgreSQL OpenAI API Claude API TensorFlow AWS Redis Kubernetes RAG Pipelines Vector DBs
From The Lab

Insights & Thinking

Practical perspectives on AI, cybersecurity, and building products that actually work — written from the trenches.

Prompt Engineering

Why Prompt Engineering Is the Most Underrated Skill in AI

Most teams treat prompting as an afterthought. After building 6 AI products, here's what I've learned: the way you talk to an AI determines 80% of its output quality — and almost nobody is doing it right.

5 min read Feb 2025
Read Article
AI Security

The Hidden Security Risks Inside Your AI Pipeline

LLMs introduce a new class of vulnerabilities — from prompt injection to data leakage — that traditional security tools completely miss. Here's how to build AI systems that are actually production-hardened.

7 min read Jan 2025
Read Article
AI Strategy

Measuring AI Performance: Why Volume Metrics Lie

A support agent who closes 100 tickets a day looks great on paper — but are they solving problems? Here's the framework behind Mizaan and why measuring impact instead of volume changes everything.

4 min read Mar 2025
Read Article

New articles every 2 weeks — covering AI, security, and what it takes to build products that matter.

Follow on Substack
Our Vision

Our Vision & Mission

At MANNAT AI Labs, we believe that technology should be a force for good—designed to serve humanity and drive meaningful progress.

Our mission is to harness the power of artificial intelligence to address critical challenges across healthcare, education, security, and productivity—developing intelligent solutions that are not only cutting-edge but also accessible, ethical, and impactful.

Every project we undertake reflects our commitment to democratizing AI—closing the gap between innovation and real-world application. We're not just building products; we're building a platform for equity, efficiency, and empowerment in the digital age.

Our vision is to create a future where AI works for everyone—fueling sustainable growth, enabling smarter decisions, and shaping a more secure, inclusive, and intelligent world.

Social Impact Focus
Open Source Core
Accessible Innovation
Global Scalability
Mohammed Jawed — Founder of MANNAT AI Labs, IIT Kanpur and TU Wien alumnus
The Founder

Meet the Visionary

Jawed (MD) — Founder & Chief Innovation Officer

IIT Kanpur TU Wien Cybersecurity AI / ML DevOps

With a unique background spanning technology, cybersecurity, and systems engineering, I founded MANNAT AI Labs to bridge the gap between AI innovation and real-world challenges in healthcare, education, and digital security. As an alumnus of IIT Kanpur (Postgraduate in Cybersecurity) and TU Wien (Master's in Engineering Management) , and with hands-on experience in DevOps, cybersecurity and AI-driven systems, I bring both technical depth and a mission-driven mindset to everything I build.

MANNAT AI Labs represents not just years of applied research and solution development—but a bold vision to build AI-powered products that are practical, scalable, and commercially viable, while staying true to ethical and human-centered design principles. Our goal is to create impactful technology that delivers real value across industries and communities.

Why Mannat AI Labs? The name "Mannat"—meaning sacred wish or vow—reflects our promise to use AI ethically, purposefully, and for the greater good. Each initiative we undertake is guided by this commitment: blending AI with purpose, and technology with promise, to build a future where innovation empowers, includes, and uplifts.

Contact

Reach Out

Questions, feedback, or interest in any of the open-source products? Drop a line.

Email github.com/JawedCIA linkedin.com/in/jawedmd