Interactive course · 6 sections · 15 min
UK Data and AI Roles and Skills in the Current Job Market
A vendor-neutral report on the UK data and AI job market in 2026. Six role families, 25 role clusters, salary bands, skill matrices, and what employers actually hire for in tech, finance, healthcare, and the public sector.
UK Data and AI Roles and Skills in the Current Job Market
Executive summary
The current UK market for data and AI work is not a single "data scientist" market. It is a layered market made up of at least six strongly recurring families: analytics and BI, data engineering and architecture, governance and data operations, data science and applied research, ML and GenAI engineering, and product, strategy, and responsible AI leadership. That mix is consistent with current UK vacancies across tech, finance, healthcare, and public-sector employers, and with UK labour-market evidence showing that AI adoption is broadening beyond pure technology firms. In late December 2025, around a quarter of UK businesses reported already using some form of AI, and by late March 2026 nearly one in five said they planned to adopt at least one AI technology within three months.
For employers already working in AI, hiring remains difficult. DSIT's AI Labour Market Survey 2025 found that 58% of responding organisations had AI-related vacancies since 2023, and 35% had at least one hard-to-fill AI vacancy; the main reasons were lack of work experience and lack of technical skills or knowledge. The same survey found that AI professionals and AI leaders were the hardest roles to fill.
The strongest, most cross-cutting skills in the UK market are still conventional rather than exotic: SQL, Python, cloud data platforms, data modelling, ETL or ELT thinking, testing and CI/CD discipline, and the ability to communicate with non-technical stakeholders. In analytics roles, Power BI, Tableau, Looker, Excel and KPI design recur heavily. In engineering roles, Databricks, Snowflake, dbt, Spark, orchestration frameworks, AWS, Azure, GCP and platform observability recur heavily. In AI-specialist roles, the growth areas are LLMs, RAG, prompt engineering, agentic frameworks, evaluation pipelines, vector search, model monitoring, and responsible AI controls.
The official skills-gap evidence aligns with what live adverts now show. DSIT reports that 97% of surveyed AI employers identified at least one AI skills gap, with understanding AI concepts and algorithms the largest, data management emerging as a new major gap, and 57% of firms reporting a technical skills gap. DSIT also found that the use of NLP and generative text AI rose sharply, and that 57% of surveyed organisations planned to use agentic AI over the next three years.
One of the clearest current tensions in the UK market is that entry routes are broadening while specialist expectations are rising. DSIT found apprenticeships had risen from 3% of AI hires in 2020 to 19% in 2025, and that informal on-the-job AI training is now dominant. At the same time, the survey still found a strong presence of postgraduate qualifications in specialist AI roles, with 84% of responding organisations using master's-level resource and 53% using PhD-level resource, while computer science remained the most common qualification area. PwC's UK analysis points in the other direction for the broader labour market: degree requirements in high-AI-exposure postings fell from 64% in 2019 to 56% in 2024. The practical reading is that the UK market is becoming easier to enter for applied delivery roles, but harder to reach at the frontier unless candidates combine technical depth with practical experience.
Sector distribution was not specified in your brief, so no sector quota was imposed. The evidence base below deliberately includes current roles from tech, finance, healthcare, and the public sector.
Market evidence and method
This report is UK-only and is based on a targeted review of current UK-oriented vacancy evidence from Indeed UK, LinkedIn UK job pages, NHS Jobs, Civil Service Jobs, and company career pages using SmartRecruiters and Greenhouse, combined with UK official and industry evidence from ONS, DSIT, BCS, IET, PwC and The Alan Turing Institute. The employer sample was intentionally spread across tech, finance, healthcare and public-sector contexts because those sectors now show materially different skill emphases.
A few market-wide observations matter before looking at individual roles. ONS shows AI use becoming mainstream in UK business. DSIT shows AI employers struggling most with experience and advanced technical understanding rather than basic awareness. PwC shows that AI-related skills continue to grow as a share of UK postings even after the wider vacancy market softened, and that information and communication, financial services, professional services, and health and social care are all seeing stronger AI-skill demand.
A notable regional pattern also remains. In DSIT's survey, 71% of responding AI organisations were based in London, the South East, or the East of England, and graduate pathways were far more common there than elsewhere. That does not mean the rest of the UK lacks AI work, but it does mean that frontier-AI and specialist-architecture roles remain geographically concentrated compared with more general analytics and platform roles.
The sector sample used here shows distinct skill accents:
| Sector | Current UK role examples reviewed | What the sector most visibly emphasises |
|---|---|---|
| Tech | ASOS Senior AI Engineer; Deliveroo Senior ML Platform Engineer; Legal & General Analytics Engineer | LLM systems, platform tooling, experimentation, self-service analytics, reliable cloud delivery |
| Finance | JPM NLP/LLM Data Scientist Lead; Barclays Data Product Manager; CMC Markets MLOps Engineer; Arch Data Quality Lead | Governance, explainability, model risk, regulated delivery, high-trust data products, forecasting and decision support |
| Healthcare | NHS Data Scientist; Senior Population Health Data Scientist; healthcare prompt-engineering roles | Python/R/SQL, KPI frameworks, patient and population-health use cases, domain safety, trustworthy retrieval and response |
| Public sector | Civil Service Data Scientist; Planning Inspectorate Lead Data Engineer; public-sector Databricks project roles | Secure cloud delivery, reproducibility, governance, service outcomes, data quality by design, stakeholder clarity |
The most useful way to read current UK vacancies is by skill frequency tier rather than by title alone:
| Skill cluster | Frequency in the reviewed market | What it tends to attach to |
|---|---|---|
| SQL | Very high | Almost universal across analyst, BI, data engineering, analytics engineering, data science and governance roles |
| Python | Very high | Strongest in engineering, data science, ML, AI and increasingly analytics roles |
| Cloud platforms | High | AWS, Azure and GCP are normal expectations from data engineering upward; specialist AI roles add managed ML/GenAI services |
| Data modelling and pipeline design | High | Central to analytics engineering, data engineering, data architecture, BI and governance |
| BI and visual storytelling | High | Power BI, DAX, dashboarding and KPI translation dominate analyst and BI hiring |
| Statistics and experimentation | High | Core in data science, product analytics, applied science and decision-science roles |
| dbt, Snowflake, Databricks, Fabric | High | Strong current UK signature of modern analytics and platform roles |
| CI/CD, testing, Git, orchestration | High | Data engineering, MLOps and AI engineering now expect software-engineering discipline |
| Governance, metadata, lineage, quality | High and rising | Governance, steward, architect and regulated-sector roles; also increasingly required to support AI safely |
| LLMs, RAG, prompt engineering, agentic AI, evaluation | Fastest-rising | AI engineer, prompt engineer, architect, product, consultant and specialist scientist roles |
UK role taxonomy and detailed role matrix
The table below groups the current UK market into 25 recurring role clusters. Some rows aggregate near-identical titles used interchangeably in live adverts, especially where employers switch between "engineer", "scientist", "specialist" and "consultant" for similar work. Salary ranges are indicative rather than exhaustive: many live UK adverts still omit pay, and London specialist roles can sit materially above national medians. Where pay was not reliably published in the accessible sample, that is stated explicitly.
| Role cluster | Typical seniority | Common alternative titles | Indicative UK pay | Core technical stack |
|---|---|---|---|---|
| Data Analyst | Entry to mid | Junior Data Analyst; Reporting Analyst; Data & Insights Analyst | £32.75k to £56.75k benchmark; entry adverts from about £29k; specialist London roles higher | SQL, Excel, Power BI/Tableau, some Python/R, dashboarding, data cleaning |
| Product / Decision Analyst | Mid to lead | Product Analyst; Decision Scientist; Senior Product Analyst | Often £40k to £70k; senior product analytics roles can exceed this | Strong SQL, event data, funnel/cohort analysis, experimentation tooling |
| BI Analyst / BI Developer | Entry to senior | Power BI Developer; BI Developer; MI Analyst | £32.5k to £56.5k benchmark; junior Power BI roles from about £28k to £35k | Power BI, DAX, SQL, semantic modelling, reporting, KPI dashboards |
| Analytics Engineer | Mid to senior | Data Analyst Analytics Engineer; Analytics Developer | Commonly aligned to BI developer or data engineer pay bands | dbt, Snowflake/BigQuery/Fabric, SQL, semantic layers, curated datasets |
| Data Engineer | Entry to mid | Data Platform Engineer; ETL/ELT Engineer; Cloud Data Engineer | £56k to £77.75k benchmark; junior adverts lower | SQL, Python, Spark/PySpark, orchestration, ETL/ELT, AWS/Azure/GCP, warehousing |
| Senior / Lead Data Engineer | Senior to lead | Principal Data Engineer; Data Engineering Lead | Often above core DE benchmark; current adverts show about £66.6k to £95k and sometimes higher | Advanced SQL/Python, architecture decisions, performance tuning, CI/CD, mentoring, platform ownership |
| Data Architect | Senior to principal | Enterprise Data Architect; Data Solution Architect | £78.25k to £102.25k benchmark; premium London posts higher | Canonical models, integration patterns, governance standards, secure scalable structures, target-state architectures |
| Data Platform / AI Platform Architect | Principal to lead | Data & AI Architect; AI Platform Architect; Principal Data & AI Architect | Often £90k to £135k where published | Platform capabilities, interoperability, training and inference workflows, cost, resilience, security |
| Data Governance Analyst / Specialist / Steward | Entry to mid | Technical Data Governance Analyst; Associate Data Governance Analyst; Data Steward | Entry adverts from about £27.3k to £34.1k; broader specialist bands vary by sector | Data dictionary, metadata, lineage capture, ownership maps, access controls, quality monitoring, training materials |
| Data Governance Manager / Head | Manager to head | Head of Data Governance; Data Governance Lead; Governance & Quality Manager | £47.5k to £72.25k manager benchmark; heads above this | Governance operating models, ownership and stewardship frameworks, metadata, lineage, privacy, quality, catalogues |
| Data Quality / Reliability / DataOps Engineer | Entry to senior | Data Reliability Engineer; DataOps Engineer; Data Quality Lead | Often unpublished in live adverts | Observability, reliability, quality controls, rule definition, monitoring dashboards, automation, incident prevention |
| Data Scientist | Mid | Applied Data Scientist; Advanced Analytics Scientist | £56.25k to £81k benchmark | Python, SQL, notebooks, feature engineering, cloud analytics platforms |
| Senior / Lead Data Scientist | Senior to lead | Principal Data Scientist; Lead Data Scientist | Often £75k to £115k in specialist or regulated environments; principal roles higher | Production-ready modelling, experimentation leadership, model deployment oversight, regulated delivery |
| Applied Scientist / ML Scientist | Mid to staff | Machine Learning Scientist; Staff ML Scientist | Often unpublished in accessible adverts | Ranking, recommender systems, causal analysis, attribution, incrementality, deep learning where relevant |
| AI Research Scientist | Mid to principal | Research Scientist; AI Scientist; Research Engineer | Often unpublished in accessible adverts | PyTorch/JAX, research workflows, benchmarking, model innovation |
| NLP / LLM Engineer or Scientist | Mid to lead | LLM Engineer; NLP Research Engineer; LLM Data Scientist | Often £70k to £100k+ in specialist London adverts | PyTorch, Hugging Face, Transformers, fine-tuning, evaluation, RAG, agent workflows |
| Computer Vision Engineer | Mid to senior | ML Engineer Computer Vision; Vision Scientist | Specialist London adverts commonly show about £80k to £130k | OpenCV, PyTorch/TensorFlow, image processing, optimisation, edge deployment, CUDA |
| Machine Learning Engineer | Mid to principal | ML Engineer; AI/ML Engineer; Model Engineer | £60k to £95k benchmark; London specialist adverts often £80k to £100k+ | Python, APIs, model deployment, distributed services, inference systems, cloud tooling |
| MLOps / ML Platform Engineer | Mid to lead | ML Platform Engineer; ML Workflow Engineer; ML Ops Engineer | Current adverts commonly land around £75k to £95k, with higher specialist roles above that | Feature stores, model registry, training orchestration, serving, monitoring, CI/CD for ML, Kubernetes |
| AI / GenAI Engineer | Mid to senior | AI Engineer; Generative AI Engineer; Agentic AI Engineer | £50k to £90k benchmark; many London adverts sit around £70k to £100k+ | LLMs, RAG, agents, vector search, orchestration frameworks, APIs, cloud deployment, evaluation and guardrails |
| Prompt / LLM Evaluation Engineer | Mid to principal | Prompt Engineer; Prompt Designer; AI Evaluation Engineer | £62.75k to £115k benchmark; London specialist markets higher | Prompt and context engineering, evaluation frameworks, adversarial tests, RAG workflows, grounding checks, rubric design |
| AI Product Manager / Data Product Manager | Senior to principal | Technical Product Manager AI; Data Product Manager | £65k to £92k benchmark; current London adverts include about £58k to £73k and higher senior bands | Product discovery, evaluation quality, instrumentation, data modelling literacy, AI workflow prioritisation |
| AI / Data Strategy Consultant | Junior to manager | AI Consultant; Data & AI Value Strategy Consultant; Data Science Consultant | £41k to £65.5k benchmark; senior consulting roles higher | Workshops, maturity assessments, value-case modelling, client analysis, use-case discovery, governance framing |
| AI Solutions Architect | Principal to senior manager | AI Architect; AI Solution Architect; Applied AI Architect | Often £90k to £135k where published | End-to-end AI/ML/GenAI architectures, integration, platform design, explainability, security, reliability |
| Responsible AI / AI Governance Manager | Manager to head | AI Governance Manager; AI Risk Specialist; Head of AI Governance | Often not published; senior-manager and head-of-function compensation is common | Bias/fairness, explainability, privacy, controls, policy, risk assessment, governance operating models |
Overlapping skill clusters and emerging skills
The best way to understand overlap is to stop thinking in terms of titles and think instead in terms of skill clusters. The current UK market repeatedly combines five layers: data foundations, software engineering, statistical reasoning, AI-specific systems work, and governance or business translation. The matrix below shows the approximate intensity of each cluster by role family, using H for core, M for regularly useful, and L for occasional or supporting. It is a synthesis of the live vacancy sample and the official skills-gap evidence rather than a mechanical count.
| Role family | SQL | Python | Cloud / data platform | BI / storytelling | Data modelling | Statistics / experiments | Classical ML | LLM / GenAI | MLOps / CI-CD | Governance / risk | Stakeholder / domain |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Analytics and BI | H | M | M | H | M | M | L | L | L | M | H |
| Analytics engineering | H | M | H | M | H | M | L | L | M | M | H |
| Data engineering | H | H | H | L | H | L | L | M | H | M | M |
| Architecture and platforms | H | M | H | L | H | L | M | M | H | H | H |
| Governance and data ops | M | M | M | L | M | L | L | L | H | H | H |
| Data science | H | H | M | M | M | H | H | M | M | M | H |
| Applied science and specialist AI | M | H | M | L | L | H | H | H | M | M | M |
| ML engineering and MLOps | M | H | H | L | M | M | H | M | H | M | M |
| AI / GenAI engineering | M | H | H | L | M | M | M | H | H | H | M |
| Product, strategy and responsible AI | M | L | M | M | M | M | M | H | M | H | H |
Three emerging clusters stand out most clearly in the current UK market.
The first is LLM systems engineering rather than generic "prompting". Live UK adverts increasingly ask for RAG, context engineering, agentic frameworks, vector search, tool use, evaluation, adversarial prompts, guardrails and observability. That is a meaningful shift away from the 2023-style market for generic prompt fluency. DSIT's employer survey backs that up: NLP and generative text use rose sharply among AI employers, and 57% planned agentic AI adoption.
The second is governed data foundations for AI. UK employers increasingly treat data governance, lineage, quality and metadata as prerequisites for safe AI deployment rather than as back-office controls. DSIT explicitly identifies data management as a new key AI skills gap, and live roles in governance, stewardship, architecture and AI platform work repeatedly tie trusted data to AI readiness.
The third is evaluation, safety and responsible AI operations. In the current UK market, responsible AI is no longer limited to policy specialists. It appears in architect, product, prompt, consultant and governance roles as practical work around bias, transparency, explainability, privacy, model risk, safe deployment and pre-release controls. This is especially strong in finance, healthcare and public-facing services.
Two final observations are worth making.
First, the UK market still rewards experience with practical delivery more than abstract knowledge. DSIT found that the main reason hard-to-fill vacancies remained hard was lack of work experience, not simply lack of formal qualifications. That is consistent with current adverts asking for deployed systems, production ownership, stakeholder delivery and operational reliability.
Second, the market is becoming more sector-shaped. Finance roles add regulatory knowledge such as BCBS 239, Basel, underwriting or market-data context. Healthcare roles add clinical or population-health context and higher trust requirements. Public-sector roles add secure delivery, reproducibility, explainability and service-design orientation. In other words, the strongest UK candidates are increasingly "T-shaped": broad core data and AI capability, but with at least one defensible domain depth.
Six recurring role families
For a presentation-ready picture, the UK market clusters into six recurring lanes. Each lane contains a small set of role clusters that share most of their day-to-day skill base.
- Analytics and BI: Data Analyst, Product / Decision Analyst, BI Analyst / Developer, Analytics Engineer.
- Data Engineering and Architecture: Data Engineer, Senior / Lead Data Engineer, Data Architect, Data Platform / AI Platform Architect.
- Governance and Data Operations: Governance Analyst / Steward, Governance Manager / Head, Quality / Reliability / DataOps Engineer.
- Data Science and Specialist AI: Data Scientist, Senior / Lead Data Scientist, Applied Scientist / ML Scientist, AI Research Scientist, NLP / LLM Engineer, Computer Vision Engineer.
- ML and GenAI Engineering: Machine Learning Engineer, MLOps / ML Platform Engineer, AI / GenAI Engineer, Prompt / LLM Evaluation Engineer.
- Product, Strategy and Responsible AI: AI Product Manager, AI / Data Strategy Consultant, AI Solutions Architect, Responsible AI / AI Governance Manager.
The map between families and core skill clusters can be summarised as: SQL and stakeholder communication cut across almost every family; MLOps and GenAI engineering are concentrated rather than universal; governance is rising across nearly all senior roles in regulated sectors.
Open questions and limitations
This report is intentionally grounded in current, accessible vacancy text and UK reports, but there are three practical limits.
Many live UK postings do not publish salary, so the pay ranges above are strongest where recruiter benchmarks and current adverts align, and weaker where only role descriptions were visible. London also continues to skew upwards relative to national ranges.
Job-board pages, especially LinkedIn and Indeed, sometimes expose only partial text in the public snippet. Where that happened, the role matrix was based only on visible requirements, not on guessed or inferred content.
Finally, "current job market" evidence is inherently perishable. The structural findings here are robust, especially the shift toward governed data foundations, platform engineering, LLM systems work and responsible AI operations, but individual adverts will age faster than the official labour-market evidence.