In partnership with
Hey everyone,
There is a strange tension in the data world right now.
On one hand, companies are hiring more aggressively for AI, governance, and high leverage data roles than ever before.
On the other hand, entire categories of data work are slowly disappearing and absorbed by automation, orchestration frameworks, and increasingly capable AI agents.
2025 is shaping up to be both the best and the worst year to be a data professional.
It all depends on which side of the shift you are standing on.
Let’s break it down.
AI Is Eating Low-Level Data Jobs (Quietly, Quickly, and Permanently)
The first job disappearing are not complex to understand. They are tasks you used to hand to junior analysts or an off-shore contractor:
Ad-hoc SQL queries
CSV cleaning
Basic dashboards
Repetitive data validation
Manual documentation
Simple ETL debugging
A single person with ChatGPT + a semantic layer can now do the work of 3-5 junior analysts.
And companies see it. This does not mean analysts are doomed. It means the definition of an analyst is changing.
The future analyst is:
A problem framer
A systems thinker
A storyteller
A domain expert
The people who only “ran queries” or “built dashboards” are the ones at risk.
The Explosion of “AI Operations” Roles
As AI moves into the enterprise, a new category of work is emerging between engineering, products, and data:
AI Operations.
This includes:
Prompt engineering with standards
Model evaluation & monitoring
Retrieval pipeline tuning
Guardrail design
Embedding drift detection
AI workflow orchestration
Safety, compliance, and hallucination management
Companies need people who understand data, AI behavior, and business outcomes. This is not a “research” role. This is hands on operational and desperately needed.
If you have spent years building pipelines, cleaning data, or debugging workflows, this is a natural evolution.
Shoppers are adding to cart for the holidays
Over the next year, Roku predicts that 100% of the streaming audience will see ads. For growth marketers in 2026, CTV will remain an important “safe space” as AI creates widespread disruption in the search and social channels. Plus, easier access to self-serve CTV ad buying tools and targeting options will lead to a surge in locally-targeted streaming campaigns.
Read our guide to find out why growth marketers should make sure CTV is part of their 2026 media mix.
Governance + Risk Talent is Becoming the Most Valuable in the Room
For a decade, data governance was seen as optional, bureaucratic, and slow.
That era is over, AI has flipped the script.
Now, all of a sudden the below are no longer “nice to have”:
Model explainability
Data lineage
Quality thresholds
Metadata completeness
Regulatory compliance
Access controls
Bias + fairness checks
Responsible AI policies
Every company deploying AI at scale now needs:
Data stewards
Model risk managers
AI Assurance leads
Governance architects
Policy owners
Compliance aligned data leaders
If you have ever worked in:
CDE
Lineage
DQ rules
Report decomposition
Model governance
Regulatory reporting
..you are exactly the talent companies are scrambling for.
This is one of the biggest career unlocks of 2026.
The Rise of the Hybrid Data Product Manager
The data PM of the past was mostly dealing with roadmaps, requirements, Jira tickets, stakeholder alignment.
The new AI era data product manager needs:
Deep understanding of data architecture
Familiarity with LLM workflows
Ability to design internal data products
Clarity on SLAs, contracts, lineage, and dependencies
Understanding of semantic layers and knowledge graphs
Ability to measure business value of AI automation
The role is exploding because companies finally realized that you can’t build AI products without treating data itself as a product.
The hybrid data PM is becoming:
The translator
The orchestrator
The decision maker
The owner of outcomes
This is where a lot of mid-career analysts and engineers will go next.
How to Future-Proof Your Skillset (Starting This Week)
Here is the simplest way to think about it:
The work that survives is the work AI can’t easily do.
So focus your growth on four themes:
Become someone who understands systems, not tasks.
Learn data modeling, architecture patterns, lineage design, business processes. AI can’t write SQL queries. It can’t design systems (yet)
Go deep on governance, risk, and responsible AI
This is where the scarcity is. This is where the money is. Learn data contracts, quality frameworks, model monitoring, AI policy, regulatory expectations. Governance is the new differentiator.
Become AI-native in your workflows
No “dabbling”. Integrating. Make AI your query assistant, debugger, documentation writer, testing tool, data profiler, workflow optimizer. If your productivity does not go up 3-5x, then you are leaving opportunities untouched.
Build domain knowledge
The future belongs to people who understand:
Banking
Healthcare
Insurance
Supply chain
Fraud
Finance
Risk
Treasury
Consumer behavior
ChatGPT can imitate expertise. It cannot live it. Domain knowledge is the most important skill you can have in AI era.
Final thoughts
2026 won’t necessarily be a bad year for data professionals.
It is a sorting year.
Those who relied on repetitive tasks will feel the pressure.
Those who evolve into high-leverage and become AI-native operators will thrive.
Your career will move in the direction of the problems you choose to solve. So choose problems worthy solving.
If you play it right, 2026 might be the best year of your career.