Skip to content
View Mihankhahp's full-sized avatar
:electron:
isLoading
:electron:
isLoading

Block or report Mihankhahp

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
Mihankhahp/README.md
Header

Typing SVG


πŸ† Impact at a Glance

1B 10x 650+ 2
LLM tokens/user/month Latency reduction Marketplace installs Mitacs Accelerate Awards

About

I'm a Senior Software Engineer and Solution Architect specializing in Generative AI, LLM systems, and cloud-native platform engineering on AWS and GCP.

My work sits at the intersection of AI research and production engineering, taking GPT-based systems from proof-of-concept to shipped, scalable products used in real sales, marketing, and workflow automation contexts. I design full-stack, event-driven platforms, build developer-facing APIs and SDKs, and architect the infrastructure that keeps them running reliably at scale.

I'm especially focused on:

  • Productionizing LLM applications (RAG, agents, MCP, multi-agent orchestration)
  • Designing serverless, event-driven cloud architectures on AWS
  • Building developer-friendly APIs, SDKs, and CRM integrations
  • Bridging technical design with business outcomes, ROI, and operational excellence

Based in Windsor, Ontario πŸ‡¨πŸ‡¦, open to senior engineering and AI/ML roles across Canada and the US (remote or hybrid)


Credentials & Research

☁️ Verified Certifications


AWS Cloud Practitioner
Cloud fundamentals, architecture, security, and pricing

AWS Databricks Platform Architect
Lakehouse architecture, platform design, cloud data workflows

Google Project Management
Structured execution, planning, stakeholder alignment

TensorFlow Developer Specialization
Applied ML, model development, AI engineering depth

Google Data Analytics
Analytics workflows, data-driven product thinking

Databricks Fundamentals
Lakehouse and data engineering core concepts

GenAI Deployment & Monitoring
Operationalizing GenAI, observability and lifecycle control

πŸ€– Applied GenAI & LLM Specializations

πŸ… Research Recognition, Mitacs Accelerate Awards

2024, Building Trust in AI-Generated Content: Innovative Strategies for Quality and Integrity Verification

2023, Design and Evaluation of Techniques for Enhancing the Utilization of Pre-Trained Language Models (GPT-3) in Sales and Marketing through Prompt Engineering and Fine-Tuning


Tech Stack

AI / LLM / ML

Cloud / Infrastructure / Architecture

Engineering / APIs / Delivery

CRM / Business Integrations


Experience

🏒 Sr. Software Engineer, GenAI Platform Engineering

Personize.ai (Robust Choice) Β· Jul 2024 – Oct 2025 Β· Canada

Architecture & Delivery

  • Architected a production-grade, event-driven serverless platform on AWS, using Lambda, EventBridge, Step Functions, SQS/SNS, API Gateway, and DynamoDB with fully decoupled microservices, enabling ~1B LLM tokens per user per month
  • Achieved a 10x reduction in system latency by decomposing a monolithic architecture into independent microservices with asynchronous, event-driven communication
  • Designed and shipped a high-volume batch-processing REST API and private Node.js/JavaScript SDK, accelerating developer onboarding for platform consumers
  • Built and shipped a native HubSpot Marketplace app with a Custom Workflow Action, embedding the platform as a native step inside clients' existing workflows, eliminating context-switching and increasing adoption

Engineering Operations

  • Established end-to-end CI/CD pipelines via CodePipeline, CodeBuild, CodeDeploy, and GitHub Actions
  • Implemented full observability with CloudWatch monitoring, structured logging, and MTTR-reducing error handling
  • Provided technical leadership, mentoring engineers, running Agile/Scrum sprints, and driving on-time delivery

Node.js JavaScript React.js AWS Lambda EventBridge Step Functions SQS/SNS API Gateway DynamoDB CloudFormation CDK CI/CD LLM Microservices Serverless


πŸ”¬ Research Assistant, AI/ML Engineering (Mitacs Accelerate)

University of Windsor + Personize.ai Β· May 2023 – Dec 2024 Β· Windsor, Ontario

Team Lead Phase (Jun 2024 – Dec 2024)

  • Led a cross-functional research team building GPT-4-based LLM solutions for an industry partner's cloud platform
  • Delivered multiple core GenAI features into production, from research to shipped product, directly expanding the partner platform's AI capabilities
  • Applied prompt engineering, RAG, and fine-tuning techniques to improve model quality, reliability, and integration depth

Individual Contributor Phase (May 2023 – Jun 2024)

  • Transformed GPT-3 / GPT-4 research into production-ready AI features for a B2B SaaS sales and marketing platform
  • Designed scalable cloud architecture supporting large AI workloads and production-grade GenAI deployment

πŸ… 2Γ— Mitacs Accelerate Award Recipient

GPT-4 LLM RAG Prompt Engineering Fine-Tuning Generative AI Cloud Architecture B2B SaaS Research & Development


βš™οΈ AI/ML Software Engineer, Generative AI Specialization

Personize.ai (Robust Choice) Β· Jan 2023 – Apr 2023 Β· Canada

  • Owned the end-to-end product launch on Google Marketplace, from zero to a live, production-ready listing, reaching 650+ installs and establishing the company's first external distribution channel
  • Built CRM integrations with Salesforce, HubSpot, and Apollo.io using OAuth2 and REST APIs, enabling bi-directional data sync across client sales workflows
  • Engineered optimized prompts for OpenAI and Google Vertex AI LLM models, improving output consistency, quality, and relevance across client-facing workflows

Node.js GCP OpenAI Vertex AI LLM Prompt Engineering OAuth2 REST API HubSpot Salesforce Apollo.io


Featured Projects

πŸ“„ Chat-on-Documents, Local-first RAG Application

Local-first document Q&A system built with FastAPI, React (Vite), Docker, and ChromaDB.

πŸ”— View Repository β†’ Documents_Chat


πŸ”Œ Personize.ai API & Private Node.js SDK

Event-driven API product and private SDK featuring:

  • Centralized API key management and user management
  • Rate limiting, usage metering, and billing integration via Stripe
  • End-to-end observability and production error handling

🀝 Personize.ai + HubSpot Native Integration

Production HubSpot Marketplace apps enabling CRM-centered AI automation, including a Custom Workflow Action that embeds AI generation directly into HubSpot workflows, eliminating context-switching for clients.

πŸ”— View on HubSpot Marketplace β†’ Personize.ai Studio


🧠 Personize Studio, Serverless B2B SaaS Platform

Cloud-native, event-driven platform managing files, AI agents, and large-scale workflows:

  • Supports ~1B tokens per user per month
  • Processes hundreds of thousands of records in automated batch runs
  • Generates 2,000+ long-form technical outputs in under 15 minutes

πŸ“Š Personize Studio, Google Sheets Add-on

No-code batch GenAI processing directly inside Google Workspace.

πŸ”— View on Google Workspace Marketplace


Education

πŸŽ“ Master of Engineering, Electrical and Computer Engineering University of Windsor Β· Windsor, Ontario, Canada Β· Sep 2022 – Dec 2023


GitHub Stats

GitHub Stats Most Used Languages

What I Bring

GenAI Platform Engineering      β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‘  Production-grade LLM systems
Cloud Architecture (AWS/GCP)    β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‘β–‘  Serverless, event-driven, IaC
Solution Architecture            β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‘β–‘β–‘β–‘  Research β†’ POC β†’ Production
REST API & SDK Design            β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‘  Developer-facing products
CRM & Business Integrations      β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‘β–‘β–‘β–‘  HubSpot, Salesforce, Apollo.io
Technical Leadership             β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‘β–‘β–‘β–‘β–‘  Cross-functional, Agile delivery

LinkedIn GitHub

Footer

Pinned Loading

  1. Documents_Chat Documents_Chat Public

    A local-first Retrieval-Augmented Generation (RAG) app with a FastAPI backend, React (Vite) frontend, ChromaDB vector store.

    Python 1

  2. aws-image-resize-to-webp-safety-pipeline aws-image-resize-to-webp-safety-pipeline Public

    AWS CDK serverless pipeline for secure image upload, malware scanning, WebP conversion, TTL cleanup, and DynamoDB status tracking.

    JavaScript 1

  3. AWS-CDK-L1-L2-L3-static-site AWS-CDK-L1-L2-L3-static-site Public

    Build and compare the same static website architecture across AWS CDK L1, L2, and L3 constructs.

    JavaScript

  4. AWS-Instance-Inspector-ALB-ASG AWS-Instance-Inspector-ALB-ASG Public

    Instance Inspector is a live AWS Auto Scaling and Application Load Balancer visualizer. It lets you pour traffic onto two independent server fleets, watch CPU and capacity change in real time, and …

    JavaScript

  5. MyPythonJourney MyPythonJourney Public

    Python