One New York engineer verifiably shipped a customer-facing LLM agent and talked about it: Alexander Shevchenko of Ramp

Asked (summary):

Which New York–based software engineers have published blog posts, engineering articles or conference talks about building a customer-facing LLM or agent feature that went to production — covering evals, guardrails, observability, tool use or failure handling? Exclude researchers, model training or fine-tuning work, internal developer tools, founders and consultants. For each, list name, personal email, current company, title, the feature shipped, the artifact title, URL and month, and personal website.

The review crossed 83 dated production-LLM artifacts (from 93 candidates) against 14 New York identity profiles as of November 12, 2026. Exactly one person survived every cross-check on location, hands-on engineering role, exclusions, and a verifiable production artifact.

Verified match · 1 of 14 New York profiles · 2 independent source hosts

Alexander Shevchenko

Head of Ramp Labs and Applied Research, Ramp · New York
Feature shipped
Ramp Sheets — a customer-facing spreadsheet agent
Talk
“Chasing capabilities while waiting for the frontier” — youtube.com
Published
October 2026
Personal website
Personal email
Not publicly found
Engineering background
Leads Ramp Labs on hands-on applied AI and ML projects; previously a software engineer at Microsoft
evaluationguardrailstool useobservability
Profile evidence
His talk covers four of the five requested topics — evals, guardrails, tool use and observability — in one artifact. youtube.com
The profile evidence pool backs his New York base, Ramp role and engineering history with 3 corroborating sources — the most of any of the 14 candidates. aie-nyc-2026.vercel.app
No publicly listed non-employer email exists for him; the field is reported as unavailable rather than guessed. alshevchenko.com
NameEmailCompanyTitleFeatureArtifactMonthWebsite
Alexander Shevchenko Not publicly found Ramp Head of Ramp Labs and Applied Research Ramp Sheets spreadsheet agent Chasing capabilities while waiting for the frontier (youtube.com) October 2026 alshevchenko.com

Method: cross-source review as of 2026-11-12. 93 production-LLM artifact candidates scanned, 83 dated on or before 2026-11-12, crossed against 14 New York profile candidates from targeted identity searches; one exact-name match met all criteria. Exclusions applied: researchers, model training or fine-tuning work, internal developer tools, founders, consultants. Other candidates failed the location, role, or artifact cross-check and are not shown as matches. Month shown is the artifact publication month.

This report was generated automatically by Keenable SELECT at a user's request, from publicly available web sources linked herein. Keenable does not review, verify, or endorse its contents and makes no representation as to accuracy, completeness, or timeliness; AI-based extraction may contain errors. Nothing in this report is investment, legal, financial, or other professional advice. All trademarks and referenced content remain the property of their respective owners; no affiliation or endorsement is implied. To report an error, rights concern, or request removal: legal@keenable.ai.

Keenable SELECTAsk your own question
Made with Keenable SELECT