We treat every dataset as a historical document, using AI to read between the lines
without pretending the models know more than the market itself.
Our methodology and its historical roots
We design our methods to respect the long arc of market structure, using AI to scale familiar analytical habits rather than replace expert judgement or promise outcomes that sit outside our role.
Every market era leaves its own fingerprint in the data, and AI lets us read those fingerprints at a scale that older methods could not touch, provided we stay honest about what the models can and cannot infer.
From a historical perspective, the move from floor trading to electronic venues
reshaped what was visible to researchers. Hand signals and verbal quotes gave way to
time-stamped messages, then to ever richer trade and quote feeds. Today, when we
speak about AI execution strategy inference, we are building on that lineage, using
models to trace how orders likely moved through fragmented venues, interacted with
visible liquidity, and responded to changing conditions. The goal is not to
romanticise the past or overstate the present, but to place each dataset in its
proper context.
Our internal method, which we call the Execution Inference Cycle, runs through repeated rounds of data conditioning, candidate pattern detection, and historical comparison. At each stage we ask how a human expert from a previous era would have described the behaviour, and then we translate that intuition into measurable features. By anchoring our models in narratives that market practitioners recognise, we make it easier for clients to challenge assumptions, propose alternatives, and ultimately trust the parts of the output that matter for their work.
We also acknowledge the limits of any reconstruction. Trade and quote data, no matter how detailed, does not fully reveal intent, constraints, or off-venue context. That is why we stress that our outputs are hypotheses, not verdicts, and why we encourage clients to combine them with their own knowledge of mandates, workflows, and oversight frameworks. Past performance does not guarantee future results, and historical execution behaviour, however well described, remains only one lens on a complex system.
We concentrate on a specific question in financial market research, how to infer likely execution behaviour from public and private data, and we approach it with an eye on both historical precedent and present day constraints.
Where Ivvixedoefob fits in the broader research landscape
Our work at Ivvixedoefob sits in a narrow but important niche: turning complex trade and quote feeds into structured hypotheses about how orders may have been executed, then comparing those hypotheses against historical patterns and alternative explanations.
Over the years, we have seen how quickly new execution styles become normal, from the early days of electronic routing to the more recent adoption of AI-driven decision support. Each wave promised to solve old problems, yet it also introduced new forms of complexity for researchers trying to understand what actually happened in the market. By focusing on execution strategy inference, we help clients trace that evolution without getting lost in marketing language or black box metrics, keeping the emphasis on observable behaviour and documented reasoning.
We place a strong emphasis on collaboration with client teams. Rather than handing over a
sealed report, we aim to make our inference steps legible to quants, market structure
analysts, and oversight specialists who want to interrogate the details. That includes
explaining where the models are confident, where multiple narratives fit the same data, and
where additional context from the client could materially change the interpretation. Results
may vary depending on data quality, market conditions, and the questions being asked, so we
design our work to be revisited as those factors change.
Because execution research can intersect with sensitive topics such as performance assessment, regulatory review, or internal governance, we are careful about how our outputs are framed. We avoid prescriptive language about what any team should have done, and instead focus on describing the range of behaviours that are consistent with the observed data. We encourage clients to involve appropriate legal, compliance, and risk stakeholders when interpreting our findings, especially where those findings may influence policy or oversight decisions.
Why we built Ivvixedoefob
We built Ivvixedoefob around a simple idea: execution data holds more structure than most teams can see at a glance. Instead of treating trade and quote feeds as noisy by-products, we use AI to infer the execution strategies that likely produced them, giving researchers and practitioners a clearer view of how orders interact with modern market microstructure.
Our background sits at the intersection of market history, microstructure analysis, and applied machine learning. We have spent years comparing how execution behaviour evolved across venues, regimes, and technologies, and Ivvixedoefob is our way of turning that experience into a practical, transparent tool for financial market research teams.
Our philosophy on AI and market structure research
AI as an extension of disciplined observation
We believe that AI should extend, not replace, the careful observational habits that
market researchers have developed over decades. In practice, that means designing
models that can be inspected, debated, and re-run as new information arrives, rather
than locking insights inside opaque scores. We frame each inference as a hypothesis
about execution behaviour, backed by observable trade and quote patterns, so that
experienced practitioners can agree, disagree, or refine the story with their own
context and judgement.
History as a guide to present behaviour
Execution data sits at the crossroads of technology, regulation, and human decision
making, so we avoid treating it as a purely mathematical object. Our philosophy is
to weave historical context into every analysis, asking how similar patterns
appeared under earlier regimes and what changed as venues, rules, and tools evolved.
By grounding our AI execution strategy inference in that longer narrative, we help
clients avoid overfitting to a narrow time window or mistaking transient behaviours
for permanent structure.
Transparency across technical and governance lines
Because our work can influence how teams discuss performance, oversight, or market
quality, we put transparency ahead of convenience. We document data sources,
preprocessing steps, model choices, and key assumptions in plain language that
quants, legal teams, and governance committees can all read. That documentation is
not an afterthought, it is part of the product, and it reflects our view that robust
financial market research depends on methods that can be explained and challenged
from multiple perspectives.
Deliberate evolution, not constant reinvention
Markets evolve quickly, and so do analytical techniques, but institutions need stability in how they interpret findings. Our philosophy is to change carefully, comparing new approaches against established baselines and keeping a clear record of what improved and what simply shifted emphasis. By treating method changes as historical events in their own right, we help clients maintain continuity in their research, even as AI tools and data sources continue to develop.
Respecting limits and shared responsibility
We recognise that any reconstruction of execution behaviour has limits, so we design
our work to sit inside broader decision processes rather than stand alone. That
includes clear reminders that past performance does not guarantee future results,
and that inferred patterns should be considered alongside risk frameworks,
qualitative insights, and independent checks. Our role is to provide structured,
historically informed views of what might have happened in the market, not to
dictate what anyone should do next.
Who we are and how we approach financial market research
Our background sits in market microstructure research, historical case studies, and applied AI, and we combine those threads to help teams interpret execution behaviour without promising outcomes we cannot control or foresee.
We see AI for market research as another chapter in a long story that runs from paper order books to fully automated execution engines. Our role is to help teams read that story more clearly, without overstating what any model can know about intent or future behaviour.
Our values
The way we analyse execution data reflects the values we bring to financial market research, shaped by years of watching how tools, venues, and behaviours evolve.
Curiosity
Curiosity drives our work at Ivvixedoefob. We approach each dataset as a new chapter in the long history of financial markets, asking what has changed, what has stayed surprisingly familiar, and how AI can help us see those threads more clearly. That curiosity shows up in the way we design our execution strategy inference pipelines, always leaving room to test alternative explanations and to revisit earlier assumptions when fresh evidence arrives. Instead of rushing to a single neat answer, we map out a landscape of plausible behaviours, then work with clients to decide which parts matter for their particular questions. In practice, curiosity means tracing small anomalies back to their roots, comparing present day trade and quote behaviour with earlier regimes, and asking why similar patterns emerged under different technological or regulatory constraints. It also means being candid about ambiguity, showing where the data supports several narratives rather than forcing a false sense of certainty. By treating every project as an opportunity to learn something about how markets evolve, we keep our methods flexible and our interpretations grounded in observable detail, rather than in marketing promises about what AI might one day achieve.
Rigour
Rigour is how we turn open questions into dependable analysis. We build our AI execution strategy inference workflows around repeatable steps, documented decisions, and cross checks against historical data, so that another researcher could follow the same path and understand where judgement calls were made. That structure matters when our outputs feed into discussions about market quality, oversight, or internal governance, where stakeholders need to see not just the conclusion but the reasoning that led there. Our commitment to rigour extends from data conditioning through to interpretation. We pay attention to missing fields, venue quirks, and structural breaks in the time series, rather than assuming that all periods are directly comparable. We test candidate models against out of sample periods, look for cases where they fail in interesting ways, and record those limitations alongside the successes. Past performance does not guarantee future results, and rigorous analysis means acknowledging that openly while still extracting useful, carefully framed insight from the data at hand.
Collaboration
Collaboration shapes how we work with clients and with each other. Execution
research sits at the intersection of quantitative analysis, technology, legal
requirements, and operational reality, so no single viewpoint is enough. We design
our outputs so that quants, market structure specialists, compliance teams, and
governance committees can all engage with the same artefacts, each bringing their
own expertise to bear on the inferred execution patterns and historical comparisons.
Inside Ivvixedoefob, collaboration means pairing technical specialists with colleagues who
focus on market history, documentation, and client communication, ensuring that
every project benefits from multiple lenses. With clients, it means regular
checkpoints where we review preliminary inferences, invite challenges, and adjust
the analysis plan in light of new context or priorities. By treating execution
strategy inference as a shared investigation rather than a one way report, we help
organisations build internal understanding that outlasts any single project or
dataset.
Stewardship
Stewardship captures our sense of responsibility toward both data and interpretation. Financial market datasets often contain sensitive details and sit within complex regulatory frameworks, especially in jurisdictions like Ireland and the wider European Union. We handle that information with care, aligning our practices with data protection expectations and client policies, and avoiding the retention of identifiers that are not necessary for the agreed analysis. Stewardship also extends to how we frame our findings, avoiding overconfident language or implied promises about future outcomes. When we infer execution behaviour, we remember that those inferences may inform discussions about resource allocation, oversight, or public reporting. We therefore emphasise the provisional nature of any reconstruction, encourage clients to combine our work with independent checks, and clearly state that past performance does not guarantee future results. By treating both data and conclusions as things entrusted to us, rather than as raw material for abstract experiments, we aim to support healthier, more informed conversations about how modern markets actually function.