Andela Research · Emergent Role Explorer

What is your job becoming?

AI is dissolving the boundaries between tech roles. We mined 47,101 Fortune 500 job postings and watched skills bleed out of 41 traditional roles, pooling into 23 recurring clusters: candidate emergent roles. Type your job title and see where yours is going.

47,101postings analyzed
2,026skills scored
41traditional roles
23emerging roles
Step 1 · Your job title (or any skill)
Try one: DevOps Engineer · Data Analyst · Product Designer · Security Engineer · Back-End Engineer
Step 2 · Tune your signal skills — remove what you don’t do, add what you do. You can also click skills on the map below.
Why can’t I add Python or Java?

Because they don’t move. This explorer reads skill bleed — skills turning up in postings for other roles far more often than chance. Everyday staples like Python, Java, and generic SQL appear in nearly every tech posting, so they can’t distinguish a Back-End Engineer from an ML Engineer, and can’t signal where either is heading. The 442 skills tracked here are the ones doing the moving. Your everyday stack still matters — it just isn’t what’s changing.

Where you are today
Where you’re being pulled
Your bridge skills
The skill-bleed map

442 skills · 4,093 cross-role connections

Emerging (2018+) Recent (2010–17) Settled (pre-2010) Not scored Unclustered skill
Each dot is a skill, sized by posting volume; links connect skills that co-occur across different traditional roles far more often than chance. Drag to pan, scroll or pinch to zoom (labels appear as you zoom in).
How this works. From 47,101 Fortune 500 job postings we scored 2,026 skills, assigned each a “home” role, and measured skill bleed which are skills appearing together in postings whose titles belong to different traditional roles, far more often than chance. Community detection over the 442-skill bleed graph surfaced 23 recurring skill clusters: candidate emergent roles, some recent and some settled blends. Your “pull” is your skills’ total connection strength into each cluster; the percentage is a relative share across candidates, not a probability, and the top result is insensitive to the scoring weights we tested. Ubiquitous skills such as Python, Java, generic SQL appear in nearly every posting meaning they can’t really help distinguish one role from another and are therefore excluded from the graph by design. This page is a research preview, not career advice. Full method, validation, and limitations are documented in the accompanying research findings.