Open to opportunities

Vismay User

@vismayuser

Message

AI Product Manager with 4+ years shipping production AI systems that deliver measurable business outcomes.

United States

Message

What I'm looking for

I’m looking to lead product for production-grade AI agents and pipelines—focusing on reliability, observability, and measurable ROI. I want to ship multimodal or RAG-driven systems that reduce manual work and improve accuracy with strong cross-functional execution.

I’m an AI Product Manager with 4+ years shipping production AI systems end-to-end, translating complex capabilities into measurable outcomes. I’m hands-on across requirements, specs, pipeline debugging, and cross-functional alignment.

Recently, I built AuditPilot, a multimodal inspection automation platform, cutting report turnaround to same-day delivery while maintaining 95%+ accuracy. I defined requirements for multimodal processing, confidence-based human review queues, and model fallback strategies—driving a 3x increase in inspections completed per inspector per week.

I also shipped an LLM compliance workflow deployed via Microsoft Copilot Studio, reducing compliance remediation time by ~70% by enabling natural-language querying, override triggers, and production write-back. Alongside this, I built RAG pipelines with evaluation and observability in mind, so the system stays reliable as teams scale.

Before that, I led data and ML product delivery—improving forecast accuracy, reducing inventory costs, cutting processing time, and building monitoring and alerting systems that reduced downtime by 70%. I’m currently developing TraceSieve to turn agent observability signals into deterministic regression tests and proactive reliability metrics.

Experience

Work history, roles, and key accomplishments

Product Data Analyst

Dash Solutions

Oct 2024 - Feb 2026 (1 year 4 months)

Architected a multi-step LangChain agent with tool-calling and database write-back to help compliance teams answer agreement-inheritance questions and apply overrides, reducing agreement remediation time by ~70%. Built a RAG pipeline on Azure AI Search with Azure OpenAI and deployed via Microsoft Copilot Studio; also owned a Power BI self-service roadmap that reduced manual reporting effort by 60%

Lang Chain Azure AI Search Azure OpenAI Microsoft Copilot azure monitor Power BI Azure Data Factory Prompt Engineering SQL

Research Assistant

Stevens Institute of Technology

Sep 2023 - Apr 2024 (7 months)

Analyzed 998 synthetic face images with DLIB/DeepFaceAI to quantify 34 distinct first impressions from 1M+ human judgments. Built LASSO/Ridge regression models to predict impression traits with 0.40 out-of-sample R-squared and RMSE of 11.84 across 34 traits.

Dlib DLIB DeepFaceAI Machine Learning I Squared Data Analysis Python

Product Manager

Kalika Steel

May 2023 - Aug 2023 (3 months)

Owned delivery of a demand forecasting system on 1.2M production records using ARIMA and regression, improving forecast accuracy by 15% and reducing inventory costs by $10M annually. Launched an ML yield optimization workflow that cut grade analysis time by 25% and increased premium Fe 500D output by 3%, and automated daily processing from 4 hours to 48 minutes.

ARIMA Regression Modeling Demand Forecasting Production Analytics Inventory Optimization

Product Manager

Infonyx

Aug 2020 - Aug 2022 (2 years)

Drove a Flybuys data processing overhaul across 10M+ records, defining pipeline requirements and reducing query response time by 40%. Launched customer behavior analytics that increased targeted offer accuracy by 25% and delivered $1.2M partner revenue uplift, and shipped real-time pipeline monitoring/alerting across 50+ retail locations to reduce downtime by 70%.

Data Pipelines Monitoring Alerting Python DataBase Systems