Who we are and why we focus on AI shock propagation
Our story is simple: we wanted tools that respect uncertainty, support better questions, and still fit within the real constraints of financial market research teams.
We built Polsolzuent for teams who feel the gap between simple scenario spreadsheets and
opaque black-box models, especially when thinking about macro shocks and their potential
market impact.
Many of us have sat in rooms where a macro announcement landed, screens lit up, and everyone reached for different tools that did not quite agree. The conversation shifted from understanding the situation to arguing about whose model was less wrong. That experience shaped how we now design AI shock propagation models: as tools that make disagreements visible and discussable, not as devices that silence them.
When we talk about macro factor shock scenarios, we are talking about carefully defined
stress events, not dramatic headlines. We work with you to pin down what is moving, how
large the move is, and which parts of the system you are most concerned about. Then we
build and test AI models that can explore how those shocks might pass through different
channels, always keeping the assumptions on the table.
We know that every organisation has its own data quirks, governance rules, and risk appetite. Rather than pushing a one-size-fits-all template, we focus on aligning our methods with your existing processes. That might mean adapting to your data pipelines, mirroring your scenario formats, or documenting outputs in a way that fits your internal review steps.
We recognise that our work touches on areas where small misunderstandings can have outsized consequences. That is why we avoid promising specific outcomes and instead focus on describing what our models can and cannot reasonably say. We encourage every team we work with to treat AI outputs as one input among many, rather than a verdict.
When we present macro factor shock scenarios and their modeled transmission paths, we
highlight uncertainty bands, alternative interpretations, and data limitations. We would
rather have a slightly longer discussion upfront than a surprised reaction later. Past
performance does not guarantee future results, and results may vary depending on
context, data quality, and how the insights are used.
Our role is to help you see how shocks might propagate, not to tell you what decisions to make. We do not manage portfolios, we do not provide personal financial advice, and we do not promise specific financial outcomes. Instead, we focus on building and validating tools that make complex systems a little more legible, while keeping human judgment firmly in the loop.
Why we built Polsolzuent
Three basis points can turn a quiet trading day into a long meeting. We started Polsolzuent after seeing how small macro surprises rippled through portfolios while the tools on the table mostly looked backward. Today we focus on one thing: building AI models that trace how macro factor shocks can move through financial markets, step by step, instead of guessing from headlines.
We work like skeptical researchers, not fortune tellers. Every scenario starts with a clear question, a documented data trail, and a validation plan. Our aim is to help risk and research teams in Ireland and beyond explore “what if” questions about macro shocks with more structure, more transparency, and fewer late-night spreadsheet marathons.
Our method
Framing questions
We start by turning vague worries into specific questions. Instead of asking what happens if conditions “get worse,” we define concrete macro factor shocks, such as changes in growth expectations or shifts in funding conditions. We then assemble relevant time series, market indicators, and qualitative markers that capture how similar shocks have behaved before, always documenting sources and caveats along the way.
Tracing paths
Once the questions are clear, we build AI models that map potential transmission paths through financial markets. We combine statistical structure with machine learning tools to explore how shocks may move across rates, credit conditions, and market sentiment. Each path is treated as a scenario, not a prediction, and we keep the assumptions visible so teams can challenge them rather than accept them by default.
Testing limits
No model leaves our notebooks without a round of stress testing. We replay historical shock episodes, compare model responses with observed outcomes, and look for places where the model overreacts or underreacts. When we find gaps, we adjust the structure, not just the parameters. The goal is not perfection; it is to know where the model is informative and where human judgment still needs to carry most of the weight.
If you need to explore how a macro factor shock could move through your markets, we can help you turn that concern into a structured, testable scenario. Share your questions, your current tools, and your constraints, and we will outline how our shock propagation modeling approach might fit alongside your existing research workflows.
Our background blends macro research, quantitative modeling, and practical work with risk teams who have to explain their views to non-specialists.
Inside Polsolzuent you will find people who have worked on everything from central bank watching to data engineering. Some of us are more comfortable with time series and model diagnostics; others focus on how to turn technical results into clear narratives that can be discussed in committees and reports. We share a preference for plain language, careful caveats, and transparent methods.
We maintain a small network of collaborators, including external researchers and practitioners, who help us challenge our assumptions. When we explore new AI methods for shock propagation, we invite them to review our approach, question our choices, and point out where the models may be reading too much into noisy data. This peer challenge keeps us honest and grounded.
Because we work with sensitive financial topics, we pay close attention to data
protection and regulatory expectations in Ireland and across the European Union. We
design our workflows so that client data stays under clear control, and we avoid
collecting more than is necessary for the analytical task at hand. Our privacy practices
are documented and updated for 2026.
The people and principles behind Polsolzuent
We combine technical curiosity with a cautious, governance-aware mindset so that AI shock propagation modeling can be discussed openly, not treated as a mysterious layer between data and decisions.
How we think about AI shock propagation
When we say we build AI shock propagation models, we mean something quite specific. We care less about flashy forecasts and more about understanding how a macro factor shock might travel through different parts of the financial system. Below is how we think about our work, in plain language, without pretending that uncertainty disappears once a model is in place.
Clarifying macro shocks
We start by identifying the macro factors that matter for your questions, then define clear shock scenarios around them. That might include shifts in growth expectations, funding costs, or policy signals. We work with your existing data sources where possible and highlight where the data is thin, so nobody mistakes a narrow sample for a broad truth.
Designing transmission maps
Next we design model structures that can trace how those shocks may influence pricing, liquidity, and sentiment over time. We use AI tools to detect non-obvious relationships, but we always anchor them in economic reasoning. The models are built to show plausible transmission paths, not to promise precise point estimates.
Testing against history
We then validate the models by replaying past episodes and comparing the modeled transmission with what actually happened. When the model diverges, we treat that as a learning signal, not a failure. This loop helps us understand when the model is informative and when it should be treated with extra caution.
Supporting better dialogue
Finally, we package the results in a way that risk and research teams can discuss. That usually means clear scenario narratives, visual maps of transmission channels, and documentation of assumptions. The aim is to support better conversations, not to replace human judgment or internal governance processes.
How we work with shocks
From messy data to testable scenarios
We came from teams where a single surprise rate announcement could send everyone scrambling through models that were never designed to talk to each other. One tool handled macro factors, another handled pricing, and a third tried to stitch everything together after the fact. The result was slow, fragile, and hard to explain to anyone who was not in the room when the models were built. We wanted something more transparent and repeatable. At Polsolzuent, we focus on AI shock propagation modeling for financial market research. In practice, that means building systems that turn macro factor shocks into structured scenarios, tracing how those shocks might transmit through interest rates, liquidity conditions, and market sentiment. We treat the models as hypotheses, not crystal balls, and we spend as much time challenging them as we do building them. Our internal workflow follows a simple pattern we call Observe, Map, Test. We observe how past shocks have moved through markets, map those movements into explicit transmission channels, then test AI models against those patterns to see where they hold up and where they break. This keeps our work grounded in data while leaving room for uncertainty. We collaborate with research, risk, and strategy teams who want to explore macro scenarios without turning their process into a black box. We are based in Ireland, work with a small network of specialists, and update our methods regularly to reflect current market conditions and regulatory expectations. Past performance does not guarantee future results, and results may vary across use cases.