Reeya Patel
Research engineer at SquadProxy focused on LLM evaluation methodology, RAG pipeline infrastructure, and the statistical validity of multi-origin evaluation studies.
Eight years in ML infrastructure across two frontier labs and one large AI-adjacent startup before joining SquadProxy in early 2025.
Reeya leads methodology work at SquadProxy. The thread across her recent writing: evaluation that treats origin region as a first- class variable, not a confound. If you've read the multilingual benchmark post or the LLM evaluation use case, that's Reeya's framing.
Background
Before SquadProxy, Reeya worked on training-pipeline infrastructure at a frontier lab (2017-2023) and then led RAG systems at a production AI company (2023-2025). The specific scar tissue that produced the RAG-pipeline post was an early-2025 migration of a production index from single-origin to multi-origin collection that exposed the canonicalisation issues our proxy infrastructure for RAG pipelines post describes.
Writing on SquadProxy
- Proxies for RAG pipelines: latency, consistency, versioning
- Residential vs datacenter for AI workloads: a routing matrix
- Proxies as methodology for multilingual LLM benchmarks
- Ethical residential proxies for AI research: provenance
What she's working on
Reproducibility tooling for multi-origin evaluation — a small set of helpers that turn "we ran the eval from 10 countries" into a reproducible artifact that a reviewer can re-run a year later and get numbers that mean the same thing. Expect a writeup in Q3 2026.
Contact
Reeya reads customer questions about evaluation methodology. The best path is hello@squadproxy.com with "methodology" in the subject line.
Writing by Reeya Patel
22 Apr 2026
Ethical residential proxies for AI research: why provenance is a methodology concern
If your training-data corpus passes through a residential proxy pool whose peers didn't meaningfully consent, the provenance problem is now your provenance problem. A practical framing of why proxy-pool provenance matters for AI research specifically, and what to ask a vendor.
22 Apr 2026
Proxies as methodology for multilingual LLM benchmarks
Multilingual LLM evaluation that uses only US-cloud-origin requests under-reports regional content policy and geo-dependent response divergence. A proxy layer anchored to each benchmark language's primary country is methodology, not infrastructure.
22 Apr 2026
Residential vs datacenter proxy for AI workloads: a routing matrix
Most AI teams over-index on residential proxies and pay too much for coverage they don't need. The useful question isn't residential-vs-datacenter; it's which source class goes through which exit class. A practical routing matrix for training, RAG, and evaluation pipelines.
18 Feb 2026
Proxy infrastructure for RAG pipelines: latency, consistency, versioning
A RAG index is only as useful as its corpus is consistent and current. The proxy layer is where consistency and currency live or die. A practical guide to picking exit classes per source, handling latency under load, and versioning re-scrapes.
Ship on a proxy network you can actually call your ops team about
Real ASNs, real edge capacity, and an engineer who answers your Slack the first time.