Skip to content
View MeghaBhasker's full-sized avatar
🏠
Working from home
🏠
Working from home

Organizations

@EarthDefine

Block or report MeghaBhasker

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
MeghaBhasker/README.md

Megha Bhasker — AI Product Manager

Autonomous systems · LLM products · Developer platforms · Open to remote

Auburn Hills, MI  ·  meghabhasker.me  ·  linkedin.com/in/megha-bhasker  ·  cloudy020501@gmail.com


What I build

I ship AI systems that make decisions — and make them reliably.

For the past 3 years I've owned the full PM loop on production AI: a multi-model autonomous routing engine (5 providers, 8 self-healing fallback strategies, real-time confidence scoring), ML-assisted pipelines processing 3M+ records, and observability infrastructure tracking 300K+ monthly API calls. I define the guardrails, write the functional specs, instrument the telemetry, and iterate off real signal.

My side projects keep me close to what's actually shippable with today's models.


Impact

300K+ 3M+ 18% 60%
Monthly API calls owned Records automated Match success boost QA rework cut
40% 25% 20–35% 3
Support investigation time ↓ Trial-to-first-call conversion ↑ API latency cut LLM products shipped

Experience

EarthDefine LLC — Product Manager, AI Platform & Automated Workflows

Jun 2024 – Present · Remote

  • Designed and shipped a production autonomous routing engine — 5 geocoding providers, 8 self-healing fallback strategies, real-time confidence scoring, override escalation logic. Boosted match success 18%, cut latency 20–35%
  • Defined responsible AI guardrails — failure modes, confidence floor thresholds, escalation policies; authored functional specs as single source of truth across engineering, QA, and legal
  • Led root cause investigation of an AI model regression — identified two-layer calibration failure, authored executive incident report, produced engineering remediation spec
  • Automated ML-assisted data ingestion pipeline for 3M+ building records — days to hours, 60% less QA rework
  • Built AI product telemetry from scratch — token-scoped dashboards tracking 300K+ monthly API calls, model fallback rates, cost-per-resolution; cut support investigation time 40%
  • Rebuilt AI developer experience — playground, sandboxes, onboarding guides — increasing trial-to-first-call conversion 25%

Venba Tech — Product Manager Intern

May – Aug 2022 · Chennai, India

  • Led AI-assisted IoT platform deployments for 3 enterprise clients — Reliance, CCD, Raintree Hotels
  • Designed automated Power BI dashboards ingesting real-time IoT sensor streams for 20+ stakeholders

Customer Analytics LLC — IoT Research Intern

Dec 2021 – Apr 2022 · Chennai, India

  • Engineered automated IoT sensor pipeline measuring 4 CNC coolant parameters in real time
  • Built end-to-end data flow from sensor ingestion to AI-assisted anomaly detection and push alerting

Projects

Next.js 14 Claude API Vercel 2026

LLM-powered tool that converts plain-language feature descriptions into engineering-ready functional specs — user stories, Given/When/Then acceptance criteria, edge cases, QA scenarios, success metrics. Owned product vision, system & API design, secure Anthropic proxy, Vercel deployment, and telemetry.

React Claude API LLM Move Analysis Vercel 2026

AI chess training platform with Claude API for real-time move analysis and natural language coaching. Adaptive onboarding, puzzle engine, AI-driven accuracy scoring, mistake review. Responsible AI design (explainability + guardrails) and telemetry instrumented throughout.

ProActiveSense — AI Fitness Wearable

IMU Sensors ML Classification DFMEA Boston Scientific Mentorship 2023

Full product lifecycle for an IMU-based movement classification wearable — user research (80+ interviews), $3B TAM analysis, DFMEA risk assessment (65% RPN reduction), go-to-market strategy.

NLP Ad Campaign Intelligence

R Twitter API NLP Sentiment Analysis 2021

Automated NLP sentiment pipeline ingesting and analyzing social data across 12 ad campaigns. Surfaced AI-driven comparative performance insights via R visualizations.


Skills

AI Product: LLM integration (Claude API) · Multi-model routing & scoring · Responsible AI & guardrails · Telemetry & evaluation design · Prompt engineering · AI product roadmapping · Functional spec writing

Technical: Next.js · NestJS · REST APIs · Docker · GitHub Actions CI/CD · Terraform · Kubernetes · Redis · PostgreSQL · Python · R · SQL · System Design · API Design

Product Management: Platform & infrastructure PM · Autonomous systems · OKRs · Stakeholder alignment · Agile/Scrum · Go-to-market

Domains: LLM products · AI-integrated platforms · Geospatial AI · Developer tools · Autonomous decision systems · IoT · Enterprise SaaS


Education

M.S. Engineering Management — Northeastern University, Boston · 2022–2024 VP, Aspiring Product Managers Club · 150+ attendee product conference · GPA 3.8

B.Tech, Electronics & Communication Engineering — VIT Chennai · 2018–2022 IEEE Published Researcher: IoT-based AI Control System (IEEE Xplore, 2021) · GPA 3.6


Publications

  • IoT-based Control System for Home Automation — IEEE Xplore, 2021
  • Smart Home Security System with Arduino & IoT — IJCRT, 2020

Open to AI PM, Platform PM, and Automation PM roles. Based in Michigan, open to remote. Require H1B transfer sponsorship — straightforward petition, no lottery.

Pinned Loading

  1. bestmove bestmove Public

    JavaScript

  2. pm-spec-generator pm-spec-generator Public

    TypeScript