We Build AI Ventures.
Not Just Fund Them.
An engineer-led venture builder in Warsaw, developing applied-AI products and quantitative research tooling.
Portfolio register
Our product work and the research and prototypes of co-founder and CTO Kacper Saks, each with its current status. Open an entry for the full record.
Clashpoint
AI sales-coaching platform: call analysis, scoring and roleplay
- Sector
- Sales Tech / AI Coaching
- Status
- In development
- As of
UKenergy
Site prospecting for battery storage (BESS) across Great Britain, built on open data
- Sector
- Energy / Geospatial Analytics
- Status
- In development
- As of
Alpha Lab
A local research workstation for honest alpha research
- Sector
- Quantitative Research Tooling
- Status
- In development
- As of
Worldclass Research Org
A research organization of AI agents in which nothing self-certifies
- Sector
- AI / Research Automation
- Status
- Published research
- As of
QuantBrief
Multi-agent market-briefing system, built at a 48-hour hackathon
- Sector
- FinTech / AI Agents
- Status
- Completed
- As of
Certo ESG Compliance Agent
Local AI assistant for CSRD/ESRS gap analysis and compliance evidence
- Sector
- RegTech / ESG compliance
- Status
- Paused
- As of
GENS
Probabilistic athlete profiling from DNA and surveys
- Sector
- Sports Tech / Genomics
- Status
- Paused
- As of
Prediction Markets Alpha
Polymarket trading-bot prototype with five strategy modules
- Sector
- Prediction Markets
- Status
- Paused
- As of
X-Quant
A quant research program that learned to say no
- Sector
- Quantitative Finance
- Status
- Archived
- As of
How we build
We build ventures ourselves rather than backing them from the outside, and we hold our own claims to the standard we would expect from anyone asking for capital.
- Built in-house
- The founders design and build every project themselves, from the architecture to the tests.
- Evidence before claims
- Pre-registered tests, append-only research records and mutation-tested integrity checks. A project figure appears here only if it can be traced to its source.
- Corrections in public
- When a result does not hold, we say so on the page where it appeared. X-Quant’s withdrawn metrics are the clearest example.
Research and publications
Kacper Saks’s work on validation and research integrity shapes how we build and assess the portfolio.
The Validation Crisis in AI Forecasting: Deflating AGI Capability Forecasts with Quantitative-Finance Overfitting Tools
Kacper Saks, 2026. Self-archived preprint.
Read on kacpersaks.dev (opens in a new tab)An Executable-Gate Multi-Agent Research Organization: Artifact, Case Study, and a Pre-Registered Gate-Calibration Study
Kacper Saks, 2026. Zenodo (v0.2.0), erratum v0.2.1 on GitHub.
doi.org/10.5281/zenodo.20645678 (opens in a new tab)
Talk to the founders
We are open to conversations with investors and strategic partners. Technical documentation for each project is available on request.