# Fraud Prevention Strategy Owner at Zeta

> Onsite · India · Unspecified · Other · open. Source: https://alderhq.app/jobs/lever%3Azeta%3A4ecfe087-4c3e-477f-a996-b5fd51d8cd78 (read 2026-11-06).

## Facts (observed)

- Company: [Zeta](https://alderhq.app/companies/zeta)
- Location text: Bangalore - EC
- Where it can be done from (inferred): Onsite · India (evidence: "Location lists "bangalore"")
- Posted: 2026-11-06 (0 days ago)
- First seen by alder: 2026-11-06 · last verified: 2026-11-06
- Reposted: 0 times · edited: 0 times
- Apply on the employer's site: https://jobs.lever.co/zeta/4ecfe087-4c3e-477f-a996-b5fd51d8cd78
- Skills listed (inferred): Python, Ruby/Rails, SQL, LLMs

## Pay and sponsorship

- Posted pay: not posted
- Market band for similar posted roles (Bengaluru, n 9): p25 ₹3L, median ₹3.6L, p75 ₹6.2L
- Licensed UK sponsor: registered as "ZETA (IOM) Limited" (checked 2026-11-03, [register](https://www.gov.uk/government/publications/register-of-licensed-sponsors-workers))

## Description (the employer's text)

### Responsibilities

You are responsible for mitigating fraud across all outgoing payment rails, including card payments and money transfers.

### Rail

### Channels

### Typologies in scope

Card rails

Debit card on Mastercard and Shazam, ATM, digital wallets

Card-not-present, counterfeit and skimming, lost/stolen, card testing, ATM cash-out, wallet token fraud, stand-in exposure

Money movement rails

ACH (outbound and inbound), P2P claim-link sends, push and pull via card rails

Unauthorized transfers, authorized push payment scams, ACH return and kiting abuse, beneficiary-add fraud, balance drain

Both

Account opening, login, profile changes

Account takeover, synthetic identity, first-party fraud, mule accounts used to cash out fraud

### Rule lifecycle

Own the full rule book on our fraud risk management (FRM) platforms

Respond to an emerging fraud pattern within 24 hours of detection: diagnose, deploy a containment rule, and document the change.

Run every rule change through a documented process: hypothesis, back-test or shadow mode, expected alert volume, approval, deploy, post-change review.

Keep a rule inventory with owner, rationale, last-tuned date and performance for each rule

Own the bank's rule approval workflow and change log

Analytics and models

Monitor transaction, login and payment data daily to spot new attack patterns before losses build.

Use SQL/Python to size losses, back-test rules and measure the false-positive cost of every control.

Apply ML and AI tools (anomaly detection, clustering, LLM-assisted alert triage)

### Operations

Prioritise the fraud roadmap by fraud exposure, friction to good customers and operational cost, and defend the trade-offs with data.

Partner with the operations team on the disputes related to fraud

Own the rule playbooks in the fraud analyst SOP and close the feedback loop from analyst outcomes into rule tuning; the Fraud Ops Lead owns the queue, SLAs and card actions.

Automate repetitive review and reporting work so analyst time goes to judgment calls.

Lead fraud incident response: contain, quantify, communicate, and write the post-mortem.

Feed confirmed fraud and dispute outcomes into Reg E dispute handling and chargeback recovery.

### Experience and Qualifications

6+ years in fraud strategy or fraud analytics at a US bank, card issuer, aquirer, neobank or payments company.

Hands-on rule authoring in a production fraud rules engine, vendor (Featurespace, Feedzai, SAS Fraud Management, FICO Falcon) or in-house.

Proven track record cutting fraud losses without a matching rise in false positives, with numbers to show it.

Strong SQL and working Python; comfortable building your own back-tests and dashboards without waiting on a data team.

Working knowledge of US payment rails and their fraud patterns: card networks, ACH (Nacha rules), and real-time or P2P payments.

Working knowledge of Reg E error-resolution obligations for card and electronic fund transfers.

Practical use of ML or AI in fraud: you have shipped, tuned or retired a model-driven control, not just read about one. You also use AI tools daily to automate analysis and manual workflows.

---
Data: alder (https://alderhq.app). Cite https://alderhq.app/jobs/lever%3Azeta%3A4ecfe087-4c3e-477f-a996-b5fd51d8cd78 and the date you read it. Observed facts come from the employer's own listing; location reach, level, family and skills are inferences.