Before you tune the AI model, inspect the data
Model changes are easy to see. In practice, inconsistent labels, weak provenance, and missing domain definitions are often the real ceiling.
Read insightDATA QUALITY FOR OPERATIONAL TEAMS
Know exactly why your data cannot be trusted.
A focused audit of CSV, Excel, database, and geospatial data — showing what is wrong, why it matters, and what to fix first.
Founder-led. Starting from €1,500. No file upload required to start.
INPUTS
CSV, Excel, database exports
DATA TYPES
Business and geospatial
DELIVERY
Typically 7–10 business days
STARTING PRICE
€1,500
WHEN TO CALL
The same cleanup keeps coming back.
A spreadsheet fix solves today's symptom. An audit finds the broken rule, source, or handoff that makes the problem repeat.
- Reports disagree across teams or systems
- Exports need manual cleanup every week or month
- Duplicates, missing fields, or invalid locations reach production
- Nobody can explain where the numbers first diverged
THE ENGAGEMENT
A decision-ready audit, not a slide deck.
You receive evidence, reproducible checks, a sample fix, and a ranked implementation plan.
What you receive
- 01Data profile and scorecardMissingness, duplicates, formats, relationships, and domain-specific checks.
- 02Business-impact mapWhich workflow fails, how often, and what the failure costs in time or risk.
- 03Sample remediationA tested cleanup on an agreed subset with rules your team can reuse.
- 04Prioritized roadmapWhat to fix now, automate next, monitor later, or accept deliberately.
FIXED ENTRY SCOPE
Starting from €1,500
Most audits are delivered within 7–10 business days after scope confirmation and secure data access.
Request an Audit- 1ScopeWorkflow, data, risk
- 2InspectProfile and reproduce
- 3PrioritizeImpact and effort
- 4ReadoutDecisions and next steps
WORK & INSIGHTS
Real projects, research, and the lessons behind them.
Client work, independent builds, and research now live in one place—with clear evidence boundaries and no duplicated case-study layer.
Extracting age from an Instagram bio is a context problem
A real client project started with a deceptively simple request: recognize “25 yo” or a birth year without treating “married in 2018” as someone's age.
Read insight
What medical AI work taught me about domain expertise
A research collaboration with physicians explored deep learning in breast cancer histopathology. The lasting lesson was how much the model depends on clinical definitions, data provenance, and honest evaluation.
Read insight
What my UAV master’s thesis taught me about data quality
My 2021 thesis combined 2D and 3D segmentation, scene comparison, geocoding and point clouds. Its most durable lesson is that model quality depends on the assumptions hidden in the data.
Read insight
Automating a spreadsheet that came back every month
A client had an Excel file that was updated every month and had to match a fixed target format. The useful part was turning that visual target into repeatable rules.
Read insightAnomaly Detection Learning Platform: start with the use case
A real client asked for a clear Google Colab demo of several anomaly detection algorithms. The hard part was not calling fit(); it was making the data and the result understandable.
Read insightWHY MESH
You work directly with the person doing the analysis.
MESH Data Solutions is a B2B data and software practice led by Mehmed Kadrić, with 8+ years across data science, custom software, data quality, geospatial data, and production systems.
Direct Founder-led from discovery through readout
Scoped Written deliverables, timeline, and security plan
Honest No oversized team claims or invented outcomes
CLIENT FEEDBACK
What clients say
Selected feedback from completed client engagements.
· Client review
thanks for the great job again.
· Client review
Mehmed did a great job. The code he wrote worked well and was clear to understand. When we needed modification, Mehmed was very attentive, quickly understood what was needed and executed on that fast. I really enjoyed working with him.
· Client review
I had the privilege of working with Mehmed on a recent project and his performance was nothing short of exemplary. His expertise, punctuality, and innovative thinking shone through in every aspect of his work. He mainta… See full review Show less
I had the privilege of working with Mehmed on a recent project and his performance was nothing short of exemplary. His expertise, punctuality, and innovative thinking shone through in every aspect of his work. He maintained clear, timely communication and his commitment to exceeding expectations was evident. Mehmed's results-oriented approach not only delivered top-quality work, but it also far surpassed my initial expectations. Looking forward to collaborating again in the future, I would highly recommend Mehmed for any project requiring in-depth knowledge, skill, and exceptional work ethic!
Original public reviews are available for verification on Upwork. Verify source
FAQ
Frequently asked questions
What kinds of data can you audit?
The audit can cover CSV, Excel, database exports, and other structured datasets across customer, sales, order, inventory, operational, and geospatial workflows. This includes addresses, coordinates, geometries, POIs, and geocoding-result quality. The final scope depends on the workflow, volume, sensitivity, and number of sources.
Do I upload files through the website?
No. The request form collects business context only. If the engagement is a fit, we agree on confidentiality and a secure transfer method before any data is shared.
How much does an audit cost?
Audits start at €1,500. The final price depends on the number of sources, data volume, business rules, sensitivity, and timeline.
How long does it take?
A focused audit is typically delivered within 7-10 business days after scope confirmation and secure access to the agreed data.
Will you clean the entire dataset?
The audit includes a sample remediation and a prioritized cleanup plan. Full production cleanup or automation is scoped separately so the audit stays fast and focused.
Can you work under an NDA?
Yes. Confidentiality and data-handling requirements are confirmed before data transfer.
Do you work with sensitive or regulated data?
Only after a separate risk, legal, and security review. The default is to use the minimum data necessary and prefer anonymized or pseudonymized samples.
NEXT STEP
Tell me which data workflow you no longer trust.
I will reply within two business days. Do not send data until scope, confidentiality, and transfer are agreed.