Responsibility

He owns the overmatch rule for patient records, and other teams build to his sign-off

As Senior Cloud Engineer at CVS Health, Saiteja is the specialist for the overmatch path and for real-time demographic events. He makes the match rule, reviews designs that touch that path before they proceed, and remains accountable for the rule. Before that, at PayPal, he was a code owner of the Pay Later decisioning service in seven live markets. Outside CVS Health he is recognised internationally in his field, artificial intelligence and data systems for regulated industries, as an IEEE conference keynote speaker, the sole or first author of four papers in that field, and a committee member for eight international conferences.

His authority at a glance

Makes
The match rule for the overmatch path, and the enterprise event taxonomy
Signs off
Other teams' designs where they touch the overmatch and demographic-events path
Accountable for
The rule that a high-confidence match may complete automatically and a low-confidence match must go to a person
Recognised for
Artificial intelligence and data systems for regulated industries, with patient identity as his specialism: an IEEE conference keynote, four papers as sole or first author, and eight international programme committees
10+
teams build on the event taxonomy he defined
14
lines of business receive a demographic change in seconds
About 40%
less manual overmatch work with the resolver and match rule he designed
7
live markets where his idempotency sign-off held at PayPal

Figures from the CVS Health patient-identity platform and the PayPal Pay Later service. CVS Health serves more than 185 million people; that is the reach of the organisation, not a personal count.

Why the decision carries weight

A wrong identity decision in a national health organisation puts one patient's record in front of a clinician who is treating someone else. Other teams depend on the person who owns that decision. If the rule is delayed or signed off badly, the wrong record can sit in production.

At PayPal the stakes were financial: a purchase that should be refused can leave a lender exposed, and a retried charge can bill a customer twice.

His remit

Since October 2024 he has been the specialist for the identity-resolution and record-reconciliation path, in particular the overmatch path and real-time demographic events, and the lead designer of both systems on that path. He makes the technical decisions other teams follow, and he signs off their changes where they touch that path.

From July 2021 to October 2024, as Software Engineer 2 at PayPal, he was the specialist for live Pay Later decisions: a code owner of the service and responsible for its technical decisions.

Decisions he makes and signs off

Each decision below was his to make or approve, at CVS Health and before that at PayPal.

DecisionHis authorityWho builds to itResult
The match ruleHe made it and remains accountable for itThe overmatch resolver and the teams that build on that pathThe business no longer waits for a person to untangle an overmatch after the fact. Manual overmatch work fell by about 40 percent, and clinicians and pharmacists can rely on the identity at the point of care.
The event design and taxonomy for demographic changesHe wrote the design proposal and took it to the architecture review board, where it became the required enterprise standardMore than ten teams, including teams outside his reporting lineA demographic change has one meaning across the platform, and reaches 14 lines of business in seconds instead of hours, at about 150 transactions per second.
Designs that touch the pathThey do not proceed until he has reviewed them against the required standardTeams that do not report to himChanges from teams outside his reporting line reach the path only after his review against the standard.
Idempotency controls, PayPalHe implemented the controls and signed them off as a code owner of the Pay Later serviceEngineers taking changes live on that serviceA timed-out retry returned the original charge instead of creating a second one, in all seven markets.
Lawful maximum capture, PayPalHe delivered the calculation and signed off the amount on the live pathEngineers releasing to the service in the United Kingdom, France, Germany, Spain, and ItalyThe live path applied the lawful maximum capture amount in each of those five markets.

How a change reaches the path

Every design that touches the overmatch or demographic-events path passes through his review before it is built.

  1. 1Another team proposes a designThe design touches the overmatch or demographic-events path.
  2. 2Saiteja reviews and signs it offIt is checked against the standard and does not proceed until he has reviewed it.
  3. 3The team builds to his decisionWork starts only after his review.

Only where a decision sets the standard: he takes it to the architecture review board, where it becomes the required standard that later designs are reviewed against.

How he leads

  • Runs working sessions so teams outside his reporting line use one meaning for a demographic change
  • Owns a shared integration library that other teams use to support data from 14 lines of business, and wrote the guide four teams follow
  • Writes technical design documents and leads architecture review sessions
  • Mentors engineers and interviews candidates for a globally distributed team

PayPal: code owner, July 2021 to October 2024

On a team of about five that owned the Pay Later service, he was responsible for its technical decisions. The service decided in real time whether a purchase was approved, captured, refunded, or disputed. No change went live without a code owner's review. He reviewed and approved other engineers' changes and took production releases live. He signed off the idempotency controls, which held in all seven markets, and the lawful maximum capture amount on the live path in the United Kingdom, France, Germany, Spain, and Italy.

The rules he signed off, market by marketPay Later service at PayPal, July 2021 to October 2024
USAUUKFRDEESITIdempotency controlsLawful maximum capture

United States, Australia, United Kingdom, France, Germany, Spain, Italy. A filled square marks a market where the rule he signed off applied on the live path. Lawful maximum capture applied only in the five European markets.

Recognised internationally in his field

His field is artificial intelligence and data systems for regulated industries, where a wrong record or a duplicate charge has real cost. Patient identity is his specialism within it, and the subject of two of his sole-author papers. Both are on the same problem as the overmatch path: linking each record to the right patient.

Peer-reviewed work in the field

  • Sole author, synthetic patient data for identity matching and deduplication (International Journal of Computer Applications, 2026)
  • Sole author, privacy-preserving patient record linkage (IEEE INTCEC 2026, oral presentation)
  • Sole author, real-time anomaly detection for fraud prevention in high-volume financial systems (IEEE UEMCON 2026, presented in person)
  • First author, data quality and observability for healthcare machine learning pipelines (IEEE HealthCom 2026)

Selected by the international profession

  • Keynote speaker at AIC 2026, the IEEE 5th World Conference on Applied Intelligence and Computing, on artificial intelligence in regulated industries: "Building AI on Generated Data Without Exposing Sensitive Information"
  • Technical or programme committee member for eight international conferences: the IEEE World AI IoT Congress (2026, and again in 2027), IEEE IHTC 2026, ICAIE 2026, ICAIES 2027, MIWAI 2026, CSEAI 2026, ISGAIE 2026, and BDAA 2026
  • Reviewer of 106 papers across 27 international conferences, workshops, and journals, and a reviewer for two IEEE journal review pools
  • Interviewed by Authority Magazine for its Guardians of AI series, on keeping artificial intelligence safe
  • Accepted an invited panel at The AI Summit New York, December 2026, on operating data platforms at scale

Professional standing

  • Elevated to IEEE Senior Member in 2026, and a Fellow of the Royal Statistical Society

Details and links are on the Public Influencer page.

Roles by year

Where he signed off other people's technical decisionsRoles by year. Blue marks the roles where he signed off other engineers' or teams' changes.
20192021202320252027Northwest Missouri State UniversityGraduate Teaching AssistantSRSoft Inc, client E*TRADESoftware Development EngineerPayPalSoftware Engineer 2; code owner, Pay LaterCVS HealthSenior Cloud Engineer; overmatch and demographic-events sign-off

Dates from his CV. The CVS Health role continues to the present.

Saiteja presenting his sole-author paper on fraud detection at IEEE UEMCON 2026
Presenting his sole-author paper on real-time fraud detection at IEEE UEMCON 2026.
Saiteja presenting his sole-author paper on patient record linkage at IEEE INTCEC 2026
Presenting his sole-author paper on privacy-preserving patient record linkage at IEEE INTCEC 2026.
Keynote certificate from AIC 2026, signed by the three General Chairs
Keynote certificate from AIC 2026, signed by the conference's three General Chairs.

Across two employers he has owned the decision a critical system must get right. At CVS Health, teams that do not report to him build to decisions he signed off, and an uncertain patient match is no longer written as certain. At PayPal, the idempotency rule he signed off held in all seven markets, and he signed off the lawful maximum capture amount in five of them.