Credit Analyst

Title:  Credit Analyst

Function:  Data

Salary:  £50k-£70k

About the company

Noggin HQ (Noggin) is a 6-person start-up based in the Newcastle upon Tyne. Having recently secured our Credit Referencing and AISP permissions, as well as a new round of institutional funding, we are looking for an exceptional Credit Analyst to support our next phase of growth. Our core products are built using Open Banking data, e.g., transaction categorisation, income verification models and credit risk scores, which exist to serve our core mission: making access to credit equitable. We are early stage with big, exciting challenges to tackle as a team.

How we work

We are a half in-person / half remote team with offices in Ouseburn (Newcastle upon Tyne). Our working environment suits people who are proactive, are comfortable working with processes that are in their infancy, own their area(s) of responsibility and can push things forward, make their work visible, and have a clear passion for what they do. For this role, flexible working arrangements can be agreed up-front.

As the only Credit Analyst in the team, you’ll own the analytical process end-to-end, from design to execution to delivery (communicating results). You'll move between deeply technical work (model validation, backtesting, monitoring frameworks) and commercial communication (presenting to prospective lender clients, reporting to the board). And will have a genuine influence on strategy.

Key responsibilities

• Lead the retrospective ("retro") analysis that we conduct to evidence the predictive value of our data products against a lender's historical loan book and other performance measures.
• Critically evaluate, challenge, and validate credit risk scorecards and models using standard validation methodology.
• Design and operate an ongoing performance monitoring framework for our credit score and data products once in production with lending customers, including defining escalation triggers and reporting lines.
• Analyse our existing and prospective datasets to directly inform our data acquisition strategy - identifying where we have strong data coverage and where we have gaps.
• Independently query, manipulate, and analyse large datasets using SQL and Python, running analyses end to end without support from a wider analytics team.
• Communicate analytical findings clearly to lenders and prospective clients to support commercial deals, and to founders and the board internally.
• Act as a proactive first line of defence on regulatory and fairness issues, flagging potential compliance concerns to the compliance function before being asked.

Required experience

To succeed in this role, you'll need to meet the following minimum experience levels. These are non-negotiable given the level of independent ownership the role carries:

• You've independently evaluated or validated a credit risk scorecard or model built by someone else, understand its construction well enough to challenge it, and apply standard validation methodology (e.g. Gini/AUC, PSI, KS statistic).
• You have 5+ years of experience in Credit Risk in a Retail banking environment.
• You've personally designed and run a retrospective or backtesting analysis before, and can speak confidently to the methodological choices and common pitfalls involved (e.g. survivorship bias, look-ahead bias, unrepresentative samples).
• You've personally designed and/or operated an ongoing performance monitoring framework for a live model or product.
• You have strong, independent proficiency in both SQL and Python (or equivalent), and have run analyses end to end without support.
• You've presented analytical findings directly to an external client or partner, in a clear and commercially relevant way.
• You’ve produced clear, well-structured analysis of datasets, or have a strong ability to describe what information matters when assessing a new data source (population, coverage, key variables, quality issues).

Desirable experience

• Conceptual understanding of how Open Banking / alternative transactional data differs from traditional bureau data, and its main data quality and coverage challenges - hands-on experience is a bonus, not a requirement.
• A working understanding of fairness and bias testing principles for credit models (e.g. disparate impact, protected characteristics, proxy variables) and the broader regulatory context (e.g. Consumer Duty, fair lending expectations, automated decision-making rules).