Our verdict
InData Labs is an AI development, implementation and consulting firm headquartered in Nicosia, Cyprus, with offices in Vilnius, Lithuania, and Miami, Florida. Marat Karpeko, who came from the video game industry, founded it in 2014. The firm reports more than 80 employees and more than 150 delivered projects, and Clutch lists a team of 50 to 249 with a 4.9 rating from 20 reviews (InData Labs; Clutch).
Its work centers on applied data science: forecasting, recommendation, fraud detection and computer vision, and more recently retrieval-augmented generation and AI agents. It is an AWS and Databricks partner. Clients it has worked with include Flo, the period-tracking app, whose CTO wrote in a 2017 Clutch review that InData Labs built its cycle-prediction models and a data platform serving more than five million monthly users, and the GSMA, the mobile operators’ association, for which it improved a tool that classifies mobile money users by gender to study the gender gap (Clutch; InData Labs).
For a startup or mid-market company that needs models built on its own data, along with the data engineering underneath, InData Labs brings twelve years of applied machine learning work and a client record that goes back further than most firms its size.
The right buyer
- Companies that need applied data science, such as forecasting, recommendation or fraud models, alongside newer generative AI work.
- Startups and mid-market firms with a pilot budget: its site gives typical ranges of $15,000 to $50,000 for a proof of concept and $75,000 to $200,000 for mid-complexity builds.
- Teams on AWS or Databricks that want the data engineering and the models from one firm.
- Buyers who value repeat clients: Flo, a trademark search company and a social media analytics firm each describe engagements on Clutch that ran a year or more.
Reasons to look elsewhere
- A trading company CEO rated a 2023 algorithm project 3.5 on Clutch, citing weeks of delays and a cost the result did not justify.
- A manufacturing client said proposals and planning calls could be more efficient and better prepared.
- 16 of its 20 Clutch reviews date from 2023 or earlier, and several recent case studies, including its LLM analysis of Jira data, do not name the client.
- It does not say where its engineers work, and a 2017 client review describes it as based in Minsk, Belarus, so buyers with vendor-location rules should ask where their team will sit.
Scores
| Criterion | Weight | Score | Basis |
|---|---|---|---|
| Production track record | 20% | 7.6 | Flo prediction models at five million monthly users, an anti-fraud system in production for a games studio, freight-rate models for a logistics firm. |
| Technical depth | 15% | 7.7 | Data science, NLP, computer vision and pose estimation; in-house R&D center; twelve years of machine learning work. |
| Industry depth | 10% | 7.1 | Repeat work in fintech, digital health, marketing technology, logistics and gaming. |
| Delivery model and handoff | 15% | 7.0 | Discovery and proof-of-concept phases with documentation; one client saw team changes between funded phases. |
| Commercial clarity | 10% | 7.9 | Publishes typical price ranges by project size; time-and-materials and fixed-price models; $10,000 Clutch minimum. |
| Platform independence | 10% | 7.4 | AWS and Databricks partner that also delivers on Azure and across several model providers. |
| Stability and scale | 10% | 7.3 | Operating since 2014 with 80+ people and offices in Cyprus, Lithuania and the US. |
| Independent evidence | 10% | 7.2 | 20 Clutch reviews averaging 4.9, including one 3.5; most reviews are from 2023 or earlier; little press coverage. |
| Overall | 100% | 7.4 | Weighted average |
Where InData Labs fits
| Need | Fit | Why |
|---|---|---|
| Predictive models on company data | Strong | Forecasting, freight-rate, anti-fraud and cycle-prediction work for clients that are identified on Clutch. |
| Data platforms on AWS or Databricks | Good | Partner with both; data lake and ingestion projects described on Clutch. |
| Generative AI apps, RAG and agents | Good | RAG knowledge base and agent builds listed, with fewer identified clients so far. |
| Large enterprise transformation programs | Mixed | 10% of Clutch clients are enterprises; most are small and mid-market companies. |
What clients and leaders say
“InData Labs has a deep bench of people with the experience we need. They’re very instructive and help us professionally develop our side, too.”
“We're a very small organization with a limited budget, but they always treated us like our problems and our business was of utmost importance.”
“The end result was not met, and the cost was well overpriced.”