Built on evidence.
Designed for real careers.
Every AIQ Score is produced by a documented process: cited exposure research, an India-specific career knowledge graph, a deterministic scoring engine, and a fail-closed evidence check before anything is written. Our goal isn't to predict your future - it's to help you prepare for it.
One question, answered with evidence.
The AIQ Score measures how exposed your specific role and skills are to AI-driven change, and which parts of your work hold their value. It is computed from your profile, a career knowledge graph, three published exposure indices and live Indian market data - so the answer is sourced, not assumed. The question it exists to answer:
“What's the smartest next step for my career?”
Not a wrapper
around a chatbot.
A career intelligence system.
JobsAIQ runs on multiple intelligence layers working together - an India-specific knowledge graph, live market pipelines, cited exposure research and a deterministic scoring engine, with reasoning models used only on top of that evidence. That's why the output is explainable, reproducible and specific to your role.
- 01
India Career Knowledge Graph
956 mapped roles, their skills, adjacencies and decay paths - built for the Indian market, not imported from a US taxonomy.
- 02
Live market data pipelines
Job listings, hiring signals and AI-workplace news ingested continuously, so exposure reflects this month - not last year's report.
- 03
Published exposure research
Eloundou 2023, Felten AIOE 2021 and ILO WP140 2025, joined through a human-verified SOC/ISCO crosswalk of 167 mappings.
- 04
Deterministic scoring engine
Transparent rules and weights ensure results are consistent, explainable and reproducible - every time.
- Evidence first
- Explainable
- Role specific
- Privacy focused
Four stages, from cited source to your report.
Ground: Cited Exposure Crosswalk
Your resume is parsed and matched against a career knowledge graph and a human-verified exposure crosswalk - 167 hand-verified SOC/ISCO mappings carrying three published AI-exposure indices. Unmatched roles get an honest empty state, not a guess.
- 167 hand-verified SOC/ISCO mappings
- Eloundou (2023) exposure index
- Felten AIOE (2021) exposure index
- ILO Working Paper 140 (2025)
- 2,391 cited exposure rows
- 956-role knowledge graph
The engine scores. The language model only explains.
Produces every number
- Your AIQ Score and each of its component factors
- Moat, urgency, obsolescence, salary bleed and India-specific signals
- Next Best Move - the scorer is re-run against every candidate action
- Cited exposure values from the human-verified SOC/ISCO crosswalk
~2,500 lines of hand-tuned, resume-driven logic. Same input, same output - every time.
Only write the explanation
- Career Thesis, Two Futures and your plan, written under strict prompt guardrails
- Every claim passes a fail-closed evidence contract - around 24 guard and scrubber modules
- No statistic is ever generated by the model; unsourced numbers are removed
- Dated predictions are written to a prediction ledger
If a number can't be sourced, it isn't shown.
Five pillars. Cited, not assumed.
Human-verified SOC crosswalk
167 hand-verified mappings from internal roles to SOC/ISCO, carrying three published AI-exposure indices: Eloundou (2023), Felten AIOE (2021), and ILO WP140 (2025).
Career knowledge graph
956 roles × 4,316 skills × 5,272 tasks, connected by 9,106 role→skill edges and 1,828 curated plus 9,043 computed pivot edges.
Deterministic scoring engine
~2,500 lines of hand-tuned, resume-driven logic - moat, urgency, obsolescence, salary bleed - with India-specific signals layered in.
Evidence contract
A fail-closed claim/evidence layer plus ~24 guard and scrubber modules. If a number can't be sourced, it isn't shown.
Prediction ledger
Dated, falsifiable predictions written at scan time - the seed of a longitudinal career-outcome dataset.
You can check every point on your score.
Each AIQ Score ships with a visible derivation waterfall, so the reasoning is inspectable rather than assumed. You shouldn't have to trust a number without understanding where it came from.
- What influenced your score
- Where your strengths lie
- Which opportunities deserve attention
- What actions can improve your future readiness
What this methodology does not claim.
A methodology is only credible if its boundaries are stated as clearly as its outputs. JobsAIQ produces structured, sourced insight for your own decision-making. It does not do the following - by design.
- Guarantee employment
- Predict your future with certainty
- Replace career coaches or mentors
- Make hiring decisions
- Rank professionals against each other
Four principles behind every insight.
Deterministic over LLM
Any factual number comes from the deterministic engine, never an LLM guess.
Fail closed
An honest blank beats a confident guess. If it isn't sourced, it isn't shown.
Zero fabricated statistics
Every figure in your report carries a source, checked by the evidence contract.
India-first
Local salary data, local job boards, and Hindi/Hinglish signal detection built in.
Your trust matters.
We designed JobsAIQ with privacy, security, and responsible AI principles at its core.
- DPDP-aligned retention, with explicit PII consent before any scan.
- We never sell your personal information.
- Your AIQ Score is private unless you choose to share it.
- Recommendations support human decision-making - not replace it.
Live and serving scans today.
The graph, the cited crosswalk and the deterministic engine are real and in production. The learning loop is the work in front of us: unifying the risk engines into one reconciled number, closing the prediction ledger against real outcomes, and running a labelled accuracy loop.
- Unify the risk engines into one reconciled, visible number
- Close the prediction ledger against real outcomes
- Labelled accuracy loop over the resume corpus
- Graph-derived options from computed adjacency edges
- Fold market data directly into the score
- Grow curated action library and salary benchmarks
About the methodology.
No. Your resume is matched against the career knowledge graph and cited exposure crosswalk, then run through the deterministic engine alongside live market data.
We don't publish the raw scoring code, but every AIQ Score ships with a visible derivation waterfall - every point traceable to a named signal.
Retention follows DPDP-aligned rules with explicit PII consent. Your data is never sold, and your report is private unless you choose to share it.
Stated plainly: the graph, the cited crosswalk and the deterministic engine are real and in production today. Reconciling the risk engines and closing the prediction ledger against real outcomes is the work in front of us.
Better career decisions begin with better information.
Understand the methodology behind your AIQ Score.
See the signals shaping your future.
Build your next career move with confidence.