R-09 · CONSENSUS DATA TERMINAL

PILOT

Founding Machine Learning Engineer, Consensus Data Terminal

Entity resolution is the unglamorous half of this product, and the half that decides whether any of the rest of it is true.

FULL-TIME · FOUNDING · REMOTE · OVERLAPPING HOURS

Apply for this role

OPENINGS

1

One. When it is filled, this page comes down.

BASE SALARY

£230,000 to £390,000

The band, published. The offer is the same whether or not you push for more.

EQUITY

0.60% to 1.30% of the Terminal

Founding equity in the company you would actually work on.

01

The problem this role exists for

People, firms, instruments, events, and the same counterparty spelled six ways across four systems. The Terminal cannot answer anything useful until those are reconciled, and it cannot be trusted until the reconciliation is auditable by the person whose data it is.

Past resolution there is extraction over the reasoning attached to every stake in the corpus, and calibration models over a record of what people believed set against what actually happened. That last one is a genuinely novel dataset, and very few people have had the chance to model it.

More on the company itself: Consensus Data Terminal

One surface over the corpusA question typed into the Terminal is answered from three joined sources: the belief corpus, the organisation's own records, and public market data, with entity resolution and learned models in between.JOINED AT QUERY TIMEBelief corpusconviction · outcomeYour recordswhatever you holdMarket datapublic, liveCONSENSUS DATA TERMINALLIVE>what is our exposure to the Q3 capacity cut?RESOLVED ENTITIESACME LOGISTICS PLCCOUNTERPARTY412 resolved mentionsPORT OF FELIXSTOWENODEof 3 in the chainQ3 CAPACITY CUTEVENTresolved · 71% expected itMODEL OUTPUTTwo tiers of exposure, one of which nobody flagged. Confidence is stated, not implied.Illustrative. The Terminal is in early pilots.
Figure 1: One surface over the corpus. A question answered from three joined sources, with entity resolution and learned models in between. Illustrative: the Terminal is in early pilots, and the screen is a sketch of the shape rather than a product shot.

02

Your first week, and a normal Tuesday

Your first week is spent inside the corpus itself, unstructured, on a question nobody here has answered: how badly does entity resolution currently fail, and on what. A number and a taxonomy of the failure modes by the end of the fortnight is a good start.

A normal Tuesday: resolution work, an extraction model over the reasoning people wrote at the moment they staked, and a recurring conversation about auditability, because every join has to be explainable to the customer who gets blamed if it is wrong. Once a week you look at calibration, which is the part of this job that does not exist anywhere else.

WHAT YOU WOULD BE WORKING IN

Python for modelling, a columnar store and Postgres with vector search alongside it, and whichever embedding and retrieval approach survives contact with the corpus. The novel part here is the data rather than the tooling, and the tooling is yours to choose.

We do not screen on tools. Nothing above is a requirement, and a stack list is a poor proxy for whether somebody can do the work. If you have gone deep in something else and would arrive with a view about why half of this is the wrong choice, that is an argument we would like to have.

03

What you would own

Not a ticket queue and not a slice of somebody else’s roadmap. Four things with your name against them in the decision log, from the first fortnight.

01

Resolution

Entities reconciled across the corpus and whatever the customer brings with them.

02

Extraction

Structure pulled out of the reasoning people wrote at the moment they staked.

03

Calibration

Models over belief against outcome. Nobody outside this group has the data.

04

Auditability

Every join explainable to the person who will be blamed if it is wrong.

04

What we are looking for

MINIMUM

What we would need to see

  • Applied experience with entity resolution, information extraction or retrieval at a scale where the naive approach failed
  • You have shipped a model whose errors were visible to a customer, and designed for that rather than around it
  • Statistical literacy about calibration specifically, not machine learning in general
  • You can explain a model's decision to somebody non-technical who has a stake in the answer

PREFERRED

What would move you up the pile

  • Record linkage, knowledge graphs or master data management in anger
  • Work with human judgement or forecasting data, which is what this corpus is
  • Experience where a model's auditability mattered legally or commercially
  • You are interested in the corpus itself, because it is the reason to take this over a better-known job

Nothing in the preferred column is a filter, and neither list is applied before a human reads you. No degree is required for anything on this page, and we have hired people with doctorates and people with no degree at all into the same rooms. If you meet most of the minimum and not all of it, write anyway and say in the first paragraph which part you do not meet.

05

Pay, and everything around it

£230,000 to £390,000 base, plus 0.60% to 1.30% of the Terminal, reviewed upward annually without you having to ask. The band is published here rather than discovered three conversations in, and the offer is the same whether or not you negotiate, because the band is the band.

Everything else is what the whole group gets and is not negotiated per offer: a holiday floor rather than a ceiling, health and dental and vision for your family, sixteen to twenty weeks of parental leave for every parent, a funded setup, monthly paid group meetups somewhere in the world, one required funded trip abroad each year, a protected day a week on the research ventures, and no non-competes.

The full list, the salary band chart and the visa sponsorship map are on the careers page. We sponsor without qualification across seventeen jurisdictions and publish exactly which.

06

Apply for this role

Four steps: a conversation with somebody who would actually work with you, one real problem drawn from work we are genuinely doing, a working session on your solution, and a decision within a week of that session. The problem is paid if it runs long, and you keep the work either way.

Or write to careers@cnsolarlabs.com.

What you send here reaches us and no one else. We use it to reply to you, we do not add you to any list, and you can ask us to delete it at any time. See the privacy policy.