V
VivekAI
VivekAI — AI Big Data Analytics · Ver. 2
Demo — vivekai.admin / VivekAI@2026 (click to fill)
© 2026 Dawell Lifescience Private Limited
System Operational GeM Registered · Category: Artificial Intelligence / Big Data / Analytics
Enterprise AI · Big Data · Analytics — Version 2.0

Enterprise intelligence,
engineered for scale.

कृत्रिम बुद्धिमत्ता · वृहद डेटा विश्लेषण Artificial Intelligence · Big Data Analytics

VivekAI is an enterprise-grade platform to collect, integrate, process, analyze and visualize structured, semi-structured and unstructured data across an entire organization — combining machine learning, deep learning, NLP, generative AI and real-time streaming analytics to power decisions at government and enterprise scale.

Product Identification
FORM · A-01
Product
VivekAI
Description
AI Big Data Analytics
Category
AI · BIG DATA · ANALYTICS
Version
Ver. 2.0
Deployment
On-Prem · Cloud · Hybrid · Edge
Compliance
AES-256 · TLS 1.3 · OAuth 2.0 · OIDC
Availability
High-Availability · DR
Registered for Government Procurement Enterprise
§ 01 — Overview

A single platform for the full analytics lifecycle.

From ingestion pipelines and distributed processing to AI model lifecycle management (MLOps), interactive dashboards, automated reporting and secure API integration — VivekAI consolidates the tools an enterprise or public-sector organization needs to make data-driven decisions with confidence.

Built for government, public sector and large enterprise environments, the platform supports predictive, prescriptive, descriptive, diagnostic and cognitive analytics workloads on the same governed foundation.

10+
Core Functional
Modules
11
AI Model Families
Supported
15+
Data Integration
Sources
13
Industry
Verticals
07
Deployment
Options
§ 03 — Capability Layers

From raw data to decisions — five stacked layers.

Each layer is instrumented, governed and observable on its own, and composes into a single vertical pipeline from ingestion to intelligence.

L / 05

Decision Intelligence

Risk analysis, fraud detection, root-cause discovery, what-if simulation and optimization — with explainable AI (XAI) auditing every recommendation.

XAIRISKOPT
L / 04

Generative AI

Enterprise chatbot, knowledge assistant, report and content generation, code assistance and document summarization — grounded on your data.

LLMRAGCHAT
L / 03

ML / DL Platform

End-to-end model lifecycle: AutoML, feature engineering, training, validation, deployment and monitoring — governed via a central model registry.

MLOPSAUTOMLREGISTRY
L / 02

Big Data Foundation

Batch and stream processing, distributed computing, data lake management and warehouse integration for structured, semi-structured and unstructured data.

SPARKKAFKALAKE
L / 01

Ingestion & Integration

Connectors to SQL, NoSQL, ERP, CRM, IoT, SCADA, REST, SOAP, files, cloud storage, Hadoop, Kafka and MQTT — with governed pipelines and lineage.

SQLIOTAPI
AES-256
Encryption
TLS 1.3
Transport
OAuth 2.0
Authorization
OIDC
Identity
MFA
Multi-factor
Zero Trust
Architecture

Bring VivekAI into your organization.

Request a working demonstration on a representative dataset, review procurement-ready pricing, or download the product datasheet and deployment guide for evaluation.

contact@vivekai.example · +91 · Available on GeM
§ 02 — Platform · Operations

Live operations console.

Real-time view of ingestion, distributed processing and the ten functional modules running on the governed VivekAI foundation.

Cluster · VVK-ABDA-PROD
Region · India (Hyderabad / Delhi)
As of · 15 Jul 2026, 14:20 IST
Records ingested
today
4.82Cr
▲ 6.4% vs yesterday
Active data
pipelines
142
▲ 3 new
Models in
production
37
— stable
Avg query
latency
218ms
▼ 11% p95
Cluster
nodes online
96/98
2 draining
Ingestion throughputEvents / sec · last 24h
0 1.0M 00h 08h 16h 24h
Stream events (Kafka ingest)
Compute utilisation% by cluster
0 50 100 72 64 88 55 79 HYD-DC1 DEL-DC2 PUN-DC3 CHN-EDGE MUM-DR
Module status — ten functional modulesUptime · 30-day rolling
#ModuleWorkloadJobs / dayUptimeStatus
01AI Analytics EnginePredictive · prescriptive18,42099.98%Operational
02Big Data ProcessingBatch · stream (Spark/Kafka)6,90499.95%Operational
03Machine Learning PlatformAutoML · training1,28699.92%Operational
04Deep Learning FrameworkVision · forecasting74299.90%Degraded (GPU-3)
05Natural Language ProcessingIndicNLP · document AI54,11099.97%Operational
06Generative AIRAG assistant · reports9,33899.94%Operational
07Real-Time AnalyticsCEP · live dashboards99.99%Operational
08Data VisualizationDashboards · GIS100.0%Operational
09Business IntelligenceSelf-service BI · OLAP3,01599.96%Operational
10AI Decision SupportRisk · fraud · what-if2,64099.93%Operational
§ 03 — Capabilities · Model Registry

AI model families & performance.

Governed model registry spanning tabular ML, deep learning, IndicNLP and generative AI — each version tracked, validated and monitored for drift.

Registry · 37 models · 11 families
Drift alerts · 2 open
Last retrain · 13 Jul 2026
Model families
supported
11
tabular → GenAI
Mean validation
accuracy
94.2%
▲ 0.8 pt
Inference calls
per day
2.1Cr
▲ 12%
GPU hours
this month
8,640
of 12,000 quota
Validation accuracy% by model family
60 85 100 95.1 96.9 89.0 93.0 85.1 91.0 XGBoost Ensemble CNN-Sat IndicNLP Speech Forecast
Inference workload mixShare of daily calls
2.1Cr calls / day Predictive / risk — 34% NLP / document AI — 24% Vision / satellite — 18% Generative / RAG — 14% Forecasting / other — 10%
Model registryProduction tier
ModelFamilyPrimary useVer.AccuracyDrift
RiskScore-TabularXGBoostScheme leakage & fraud riskv4.295.1%Nominal
CropYield-ForecastLSTM / TransformerDistrict yield forecastingv3.091.0%Nominal
SatLandUse-CNNCNN (vision)Land-use from satellite imageryv2.789.0%Watch
DocIntel-IndicBERTIndicNLPGrievance & document triagev5.193.0%Nominal
VaaniSpeech-ASRSpeech (ASR)Hindi/regional call transcriptionv1.985.1%Watch
Anomaly-IForestIsolation ForestPDS / GST anomaly detectionv2.292.4%Nominal
SahayakRAG-AssistGenerative (RAG-LLM)Grounded policy Q&A assistantv3.4Nominal
§ 04 — Industries · Deployments

Deployments across India.

Live VivekAI programmes across central ministries, state governments and public-sector undertakings — monitoring flagship schemes at national scale.

States / UTs live · 19
Central ministries · 11
Schemes monitored · 28
States & UTs
live
19
▲ 2 onboarding
Districts
covered
612
▲ 34
Beneficiary records
under analysis
86Cr
de-duplicated
Flagship schemes
monitored
28
across 13 sectors
Top states by data volumeRecords under analysis · crore
Uttar Pradesh 13.8 Maharashtra 10.6 Bihar 9.5 West Bengal 8.2 Madhya Pradesh 7.1 Tamil Nadu 6.5 Rajasthan 5.7
Sector deploymentsFlagship programmes
SectorScheme / programmeMinistry / dept.RecordsCoverageStatus
AgriculturePM-KISAN direct benefitMoA&FW11.0 CrLive
Rural DevelopmentMGNREGA muster & wagesMoRD14.2 CrLive
HealthcareAyushman Bharat (PM-JAY)MoHFW / NHA30.1 CrLive
EducationUDISE+ school analyticsMoE1.5 CrLive
Food & PDSNFSA ration off-takeDoF&PD80.0 CrPilot
WaterJal Jeevan Mission (FHTC)DoDW&S19.4 CrPilot
BFSIGST return anomaly analyticsMoF / CBIC1.4 CrOnboarding
§ 05 — Security & Deployment

Security posture & deployment.

Zero-trust architecture with continuous compliance monitoring — deployable on-prem, in sovereign cloud, hybrid or at the edge, aligned to CERT-In and MeitY guidance.

Posture · Zero Trust
Data residency · India only
Audit · CERT-In empanelled
Platform
uptime (SLA)
99.98%
▲ within SLA
Threat events
blocked · 7d
12,840
0 breaches
MFA
coverage
100%
all privileged
Patch
compliance
98.6%
4 nodes pending
Threat events blockedPer day · last 7 days
Mon Tue Wed Thu Fri Sat Sun 0 3k
Compliance controlsContinuous audit
Encryption at rest
AES-256 · HSM-backed keys Pass
Transport
TLS 1.3 · mTLS between services Pass
Identity
OAuth 2.0 · OIDC · SAML SSO Pass
Access
RBAC + ABAC · least privilege Pass
Data residency
India-only · sovereign cloud Pass
Audit logging
Immutable · WORM retention Pass
Vuln. scanning
CERT-In cadence 4 open
Deployment optionsSeven certified topologies
/ 01 On-Premise

Government data centre

Air-gapped or restricted-network install inside NIC / state data centres with full operator control.

/ 02 Sovereign Cloud

MeitY-empanelled cloud

Deployed on GI-Cloud (MeghRaj) or empanelled providers with India-only data residency.

/ 03 Hybrid

Split control & compute

Sensitive data on-prem, burst compute to sovereign cloud with governed data movement.

/ 04 Edge

District & field nodes

Lightweight inference at district offices and field devices with store-and-forward sync.

/ 05 High-Availability

Active-active clusters

Multi-node clusters with automatic failover and zero-downtime rolling upgrades.

/ 06 Disaster Recovery

Cross-region DR

Warm-standby replica in a second region (e.g. HYD ↔ DEL) with RPO < 5 min.

§ 06 — Technology Stack

The full stack, layer by layer.

Open, standards-based components assembled into one governed platform — from ingestion connectors to the generative-AI serving layer.

Layers · 5
Connectors · 15+
Open standards · first-class
L / 05

Serving & Experience

Dashboards, GIS analytics, self-service BI, REST/GraphQL APIs and a grounded generative-AI assistant layer for policy Q&A.

REACTGRAPHQLRAG
L / 04

AI / ML Runtime

Training and inference on PyTorch and TensorFlow, served via Triton, with a central model registry, feature store and drift monitoring.

PYTORCHTRITONMLFLOW
L / 03

Processing

Distributed batch and stream computing on Spark and Flink, orchestrated by Airflow, with governed lineage across every pipeline.

SPARKFLINKAIRFLOW
L / 02

Storage

Data lake on object storage with a lakehouse table format, columnar warehouse for OLAP and vector store for retrieval.

ICEBERGPARQUETVECTOR
L / 01

Ingestion & Integration

Connectors to SQL, NoSQL, ERP, CRM, IoT/SCADA, REST, SOAP, files and cloud storage, streamed through Kafka and MQTT.

KAFKAMQTTCDC
Codebase compositionShare of services
0 40% 34% 26% 20% 12% 8% Python Java/Scala Go TypeScript SQL/Other
Environment specificationReference cluster
Orchestration
Kubernetes 1.29 · Helm
Message bus
Apache Kafka · MQTT (IoT)
Processing
Spark 3.5 · Flink 1.18
Warehouse
Columnar OLAP · Iceberg tables
Model serving
Triton · ONNX Runtime
Observability
Prometheus · Grafana · OpenTelemetry
Languages (Indic)
Hindi + 11 regional · IndicNLP
§ 07 — Pricing & Licensing

Procurement-ready licensing.

Transparent, GeM-listed tiers scaled from a single department to a national programme — annual subscription, perpetual on-prem or managed models available.

Listing · GeM · Government of India
OEM · Dawell Lifescience Pvt. Ltd.
Currency · INR · exclusive of GST
Tier I

Department

₹ 42 Lakh / yr

Single department or district. Up to 25 users, 3 modules, on-prem or sovereign cloud, standard support.

25 USERS3 MODULES8×5 SUPPORT
Tier II · Popular

Ministry / State

₹ 1.6 Cr / yr

State-wide or ministry deployment. Up to 250 users, all ten modules, HA cluster, DR, priority support.

250 USERS10 MODULESHA + DR24×7
Tier III

National Programme

Custom

National, multi-ministry roll-out. Unlimited users, dedicated tenancy, on-site engineering, custom SLAs.

UNLIMITEDDEDICATEDON-SITE
What's included by tierFeature comparison
CapabilityDepartmentMinistry / StateNational
Functional modules31010 + custom
Named users25250Unlimited
High availability & DRIncludedIncluded
Generative-AI assistantAdd-onIncludedIncluded
Support window8×524×724×7 + on-site
Deployment modelsCloud / On-premAll 6 topologiesAll 6 + air-gapped
§ 08 — Resources

Documentation, downloads & reports.

Evaluation collateral, technical documentation and generated analytics reports for procurement, deployment and operations teams.

Build · 2026.07
Docs version · Ver. 2.0
Access · Authenticated
DocumentAudiencePagesUpdatedFormat
Platform Architecture GuideSolution architects148Jul 2026PDF
Deployment & Hardening HandbookInfra / SecOps96Jul 2026PDF
Model Registry & MLOps ManualData science72Jun 2026PDF
REST / GraphQL API ReferenceIntegratorsJul 2026Web
CERT-In Compliance DossierAudit / procurement54May 2026PDF
DATASHEET

VivekAI Product Datasheet

Two-page overview of modules, deployment options and compliance — for evaluation and procurement.

WHITEPAPER

AI for Public-Sector Analytics

Reference architectures and case patterns for scheme monitoring, leakage detection and citizen services.

CHECKLIST

Procurement Readiness Checklist

GeM-aligned evaluation checklist covering security, deployment, SLAs and data-residency requirements.

SANDBOX

Evaluation Sandbox Access

Time-boxed sandbox pre-loaded with a representative synthetic scheme dataset for hands-on evaluation.

ReportScopeGeneratedScheduleStatus
Scheme Leakage — WeeklyPM-KISAN · all states14 Jul 2026Weekly · MonReady
District Yield ForecastKharif 2026 · 612 districts13 Jul 2026MonthlyReady
PDS Anomaly DigestNFSA off-take15 Jul 2026DailyRunning
Grievance Sentiment SummaryCPGRAMS · IndicNLP15 Jul 2026DailyReady
Platform SLA & SecurityCluster VVK-ABDA-PROD15 Jul 2026WeeklyReady
Logout