DATA STRATEGY · BI · AI · AUTOMATION

DATA & AI SYSTEMS
FOR MODERN TEAMS

We build data & AI systems that actually run your business. From analytics foundations to production-ready machine learning and custom internal tools, Meniva helps growing companies turn scattered data into reliable systems that drive decisions and revenue.

Book Free ConsultationView Our Work

Revenue scale supported

Annual cost reduction

Revenue uplift delivered

Saved via automation

// Production registry - 05 live deployments

Systems running in production environments.

Scope / Data - BI - ML - Agentic
Window / 2024 - 2026

01

Mi utcánk

Mi utcánk

Civic media

eu-central-1/since 2024

02

Revon

Revon

Scoring system

eu-west-1/since 2025

03

Nullfall

Nullfall

Adaptive learning

eu-central-1/since 2025

04

Scoutbound

Scoutbound

Agentic workflow

eu-central-1/since 2025

05

My Digital Inheritance

My Digital Inheritance

Data vault

eu-west-1/since 2026

What We Do

Service Modules

End-to-end data and AI capabilities, delivered as focused engagements with clear scope and measurable outcomes.

Data Foundations & Analytics

We design scalable data architectures and reporting systems that give leadership the clarity they need to act.

Data strategy & architecture
KPI frameworks & data modeling
BI dashboards (Power BI, Tableau, Looker)
SQL analytics & ad-hoc reporting
Discuss this

BI & Executive Dashboards

Interactive dashboards and self-service analytics that connect to live data and update automatically.

Power BI / Tableau / Looker builds
Executive & operational dashboards
Automated reporting pipelines
Self-service analytics layers
Discuss this

ML & Forecasting

From demand forecasting to churn prediction, we build ML systems that solve concrete business problems.

Demand & sales forecasting
Churn & retention models
Pricing optimization
Recommendation systems
MLOps pipelines
Discuss this

AI Automation & Agentic Systems

LLM-powered agents, automated workflows, and custom internal tools that replace manual busywork.

LLM & agent-based systems
Lead scoring & qualification engines
Automated data collection & ETL
Custom Streamlit / Dash apps
Internal productivity tools
Discuss this

Client Feedback

What People Say

“

“You'll absolutely love working with Bálint because he is always ready to transform data into fully accessible and actionable insights. When I worked with him on a project, I was impressed by his professional skills and attention to detail.”

C

Csaba Mezei

Communications Designer & Marketing Manager

“

“I recommend Bálint for his exceptional data analysis skills. He consistently delivers quality analysis, demonstrating thoroughness and attention to detail. His commitment to continuous improvement makes him a valuable contributor to any team.”

G

Gáspár Horváth, MBA

Founder @ Miutcank

“

“Bálint was extremely helpful and flexible from the very first moment. I could reach him anytime, he responded quickly, and always suggested up-to-date solutions. What I especially appreciated was that he didn't leave me on my own after the website was finished, but continued to support me in using it.”

A

Anna Laura Selmeci, JD

Founder & Digital Legacy Specialist

Bálint - Founder of Meniva

About Meniva

Hi, I'm Bálint

I'm a Machine Learning Engineer who loves turning messy data into business impact. I founded Meniva to help growing companies build the data infrastructure they need without the overhead of a full in-house team.

With experience spanning e-commerce analytics, sales forecasting, and AI automation, I focus on practical solutions that drive real revenue and save real time. No buzzwords, just systems that work.

5+

Years in Data & AI

20+

Projects Delivered

Connect on LinkedIn

How We Work

Our Process

A structured approach that delivers results. No surprises, just clear milestones.

01

Discovery

Free consultation to understand your data challenges, goals, and current infrastructure.

1-2 days
02

Proposal & Scope

Detailed proposal with clear scope, timeline, fixed pricing, and expected deliverables.

2-3 days
03

Build & Iterate

Agile development with weekly check-ins. We build in focused sprints with continuous feedback.

2-6 weeks
04

Deliver & Support

Production-ready delivery with documentation, training, and optional ongoing support.

Ongoing

// Release log - 04 operational systems

What we actually shipped.

Four production AI systems, each represented as a compact deployment record with live status, evaluation signals, and a direct path into the full engineering dossier.

Sorted / newest first
Window / 2025.03 - 2025.09
Filter / live systems
SEC 01/tastetrend
Type/ Retrieval
Class/ Grounded RAG
Deployed/ 2025.03
Status/ Live

Retrieval system

TasteTrend Analytics

Headline

74k+

Reviews indexed

A production retrieval-augmented generation system that transforms an unstructured corpus of 74k+ restaurant reviews into a queryable intelligence layer. Operators ask natural language questions; the system retrieves, reranks, and generates grounded answers with source citations.

M01

74k+

Reviews indexed

M02

81%

Recall@5

M03

52×

Faster analysis

Open retrieval traceOpen demo
idx 312k chunks
eval recall 81%
runtime p95 220ms
SEC 02/revon
Type/ Scoring
Class/ Calibrated ML
Deployed/ 2025.07
Status/ Live

Scoring system

Revon

Headline

78%

Top-3 prioritisation accuracy

A hybrid ML scoring system combining 12+ feature families — structured firmographic signals and LLM-extracted intent features — to rank inbound leads and surface next-best-actions. Replaces manual SDR triage with a calibrated, SHAP-explained score pipeline.

M01

78%

Top-3 prioritisation accuracy

M02

1.8s

Avg inference latency

M03

€850k

Modelled pipeline uplift

Inspect scoring runOpen demo
idx 47 leads/day
eval ECE 0.04
runtime p95 1.8s
SEC 03/nullfall
Type/ Adaptive
Class/ RAG + IRT
Deployed/ 2025.06
Status/ Live

Adaptive system

Nullfall

Headline

68%

Quiz completion increase (target)

An adaptive learning platform for ML/AI engineers that replaces static course sequences with personalised progression paths. Placement via IRT-based CAT, RAG-grounded explanations (768-dim, 1,200-char chunks, top-k 8), LLM quiz generation with pedagogical quality gates, and spaced repetition — all tracked through an explicit XP system.

M01

68%

Quiz completion increase (target)

M02

768-dim

Embedding dimension

M03

Top-k 8

RAG retrieval depth

Open adaptive engineOpen demo
idx 1,224 pages
eval grounded 84%
runtime p95 1.4s
SEC 04/scoutbound
Type/ Orchestration
Class/ Agentic
Deployed/ 2025.09
Status/ Live

Orchestration system

Scoutbound

Headline

4×

Faster prospect research

An agentic prospecting system that takes a company list, inspects live websites via Playwright browser automation, extracts structured profiles with LLM assistance, enriches with firmographic signals, scores for ICP fit, and produces a CRM-ready export — running end-to-end in an avg of 2m14s per company.

M01

4×

Faster prospect research

M02

83%

Extraction consistency

M03

2m14s

Avg workflow completion

Replay workflowOpen demo
idx 500+ layouts
eval complete 91%
runtime run 2m14s

Our Stack

Technologies We Work With

Languages & Analytics

Python
SQL
PySpark
pandas
NumPy
scikit-learn

Visualization & BI

Power BI
Tableau
Looker
Plotly
Streamlit
Analytics

ML & Cloud

PyTorch
AWS
Azure
Docker
Git
Jira

Insights

Latest Articles

Practical articles on data strategy and AI for SMEs.

Why SMEs in Europe Can’t Ignore Data Strategy in 2025

Data is no longer optional — it's a requirement for growth. Learn how SMEs can stay competitive with the right strategy.

Read more →

Coming Soon: More Insights

Stay tuned for upcoming articles about business intelligence, AI adoption, and practical data strategies for SMEs.

Read more →

Questions

Frequently Asked Questions

LET'S BUILD

Ready to turn your data into a competitive advantage? Let's talk about your challenges.

Get in Touch

Location

Budapest, Hungary

Book Free Consultation