2024–2026
From initial concept through development, launch, evaluation,
and iterative improvement.
I founded and independently developed Stark AI from an initial customer problem through product development, launch, real-world usage, evaluation, and iterative improvement. The project gave me practical experience translating generative AI capabilities into a customer-facing product, rather than treating AI as a standalone technology.
0 → 1
Independently transformed a customer problem into a functional
AI-powered SaaS product.
End-to-End
Product, AI workflows, onboarding, positioning, pricing,
customer acquisition, and go-to-market.
Real Users
Validated the solution through real customer engagement,
feedback, and product observation.
Product Problem
Making professional communication faster without making it generic.
Stark AI was designed around a simple workflow problem: professional email responses take time, but generic automation can easily lose context, tone, and usefulness. The product aimed to generate personalized, context-aware responses that reduced repetitive communication effort while still producing output appropriate for the user and situation.
My role was not limited to building prompts. I translated customer needs and workflow requirements into product capabilities, designed how customers interacted with the service, evaluated output quality, and continuously adjusted both the AI workflow and the broader product experience.
Product Lifecycle
From problem discovery to iterative improvement.
01
Discover
Identified customer use cases, workflow pain points, and opportunities for AI-enabled communication.
02
Design
Defined product workflows, onboarding, customer journey, interface, and AI-enabled capabilities.
03
Build
Developed end-to-end generative AI workflows and prompt approaches for personalized responses.
04
Evaluate
Assessed responses for accuracy, consistency, context, tone, and practical business usefulness.
05
Improve
Used feedback and recurring quality issues to prioritize prompt, workflow, and product improvements.
Founder Scope
Product, AI, customers, and go-to-market.
Customer Discovery
Conducted customer discovery, usability testing, market research, and competitive analysis to understand customer needs and adoption barriers.
AI Workflow Design
Designed end-to-end workflows using large language model capabilities to automate and improve professional communication.
Product Experience
Designed the customer journey, onboarding experience, interface, and product workflows with usability and adoption in mind.
Prompt Engineering
Developed prompt approaches to improve personalization, consistency, contextual relevance, tone, and output quality.
Product Prioritization
Used customer feedback and product observations to prioritize requirements, features, and iterative improvements.
Go-to-Market
Managed product positioning, pricing, customer acquisition, website development, and go-to-market execution.
AI Quality & Iteration
Evaluating usefulness, not just whether the model produced text.
I evaluated AI-generated responses for accuracy, consistency, contextual relevance, tone, and business usefulness. When recurring issues appeared, I translated those findings into prompt changes, workflow adjustments, or broader product improvements.
This created a practical feedback loop between customer usage, output evaluation, product requirements, and AI behavior — an experience that helped me understand how product quality in generative AI depends on the whole system, not just the underlying model.
Responsible AI
Designing for useful automation with human oversight.
Output Quality
Considered reliability, contextual relevance, consistency, and appropriate use when evaluating generated responses.
Privacy Considerations
Treated privacy as part of the product and workflow design, particularly given the sensitivity of professional communication.
Human Oversight
Positioned AI-generated communication as an assistive workflow, with users maintaining responsibility for reviewing and using generated outputs appropriately.
What I Practiced
The intersection of product, business, and generative AI.
Stark AI gave me practical experience connecting AI capabilities and limitations with customer workflows, business requirements, usability, product adoption, and productivity opportunities.
AI Product Strategy
Generative AI
Product Discovery
Workflow Design
Prompt Engineering
Output Evaluation
Customer Discovery
Usability Testing
Product Prioritization
Responsible AI
Pricing & Positioning
Go-to-Market
What Stark AI demonstrates.
The most important outcome was not simply creating an AI feature. It was taking an ambiguous customer problem, turning it into a working product, putting that product in front of real users, evaluating how both the AI and the experience performed, and iterating based on what I learned. Stark AI is where my background in strategy, customer relationships, and business development began to connect directly with AI product development.