AI infrastructure engineer

Giedrius Šteimantas

Oxylabs · 2026

Giedrius Šteimantas led scraping engineering at Oxylabs at World’s Fair 2026, with responsibility for scraper API teams and infrastructure that gathers public web data. His work on shopping agents separates discovery, product evaluation and checkout into distinct tools. He emphasizes validating retrieved content before passing it to a model and reserving browser automation for tasks that need it.

1 conference talk

AI Engineer World's Fair 202615:03

The Missing Layer in Agentic AI — Giedrius Šteimantas, Oxylabs

Giedrius Šteimantas describes rebuilding a shopping agent around web-access infrastructure. He recommends search APIs for discovery, validated scraper output for product evaluation, and browser automation for checkout. His central argument is that validating content before sending it to a model and limiting browser use can reduce wasted tokens and improve…

Giedrius Šteimantas

References

ai.engineer/talks/XsvUhpnHepE-missing-layer-in-agentic-ai ↗