Founder × Intern
A high-fidelity chronicle documenting my 2-week AI/ML & Prompt Engineering residency at Virtual Height. Mapping the interface between non-deterministic generative intelligence and deterministic system integration.
Featured Logs
Selected HighlightsInternship Timeline
Days 01 - 14Day 00: Onboarding, Orientation, and the 2-Week AI Roadmap
Walking into Virtual Height for day zero: getting our badges, meeting our lead mentors, and mapping out what we will build over the next two weeks.
Day 01: First Day as an AI Intern, Python Basics, and a Wild Flex
Day 1 kicks off with lead mentor Rashmi. We break down AI vs ML vs Deep Learning, test Python terminal scripts, and demo my custom chatbot to the room.
Day 02: Python If-Else, While Loops, and a Birthday Surprise
Tackling Python if-statements, match-case, and loops. Plus, solving a chocolate challenge and surprising our mentor with a custom birthday webpage.
Day 03: Strings, My First Terminal Chatbot, and Whiteboard Duty
Exploring Python string cleaning methods, building a rule-based terminal bot in 60 seconds, and getting called to the whiteboard to teach the class.
Day 04: APIs, JSON Data, and Hooking Up Google Gemini
Breaking out of offline scripts. We learn how APIs work, parse JSON data, and build a local Flask server powered by Google's Gemini AI.
Day 05: Prompt Engineering, AI Temperature, and the RAG Secret
Closing out Week 1 with real prompt engineering: testing temperature sliders, token math, and seeing how RAG lets AI read private documents.
Weekend 01: The Corporate 9-to-5 Loop and the 'Majority Wins' Rule
Reflecting on my first week in a corporate office. Realizing the danger of the 9-to-5 burnout loop, and why your peer environment defines your ceiling.
Day 06: Ditching the Black Terminal for a Clean HTML/CSS Chat UI
Week 2 begins! We move from boring command-line terminals to a real web page using HTML and CSS, and demo a fun custom chatbot.
Day 07: Flask Backends, HTML Forms, and the 120-Line Reality Check
Connecting our HTML frontend to Python using Flask. Understanding routes, Jinja templates, and noticing the gap between classroom pace and real development.
Day 08: Hooking Up the AI Brain: Flask Meets Hugging Face APIs
The big integration day: connecting our Flask backend to real LLMs via the Hugging Face API, and skipping massive PyTorch downloads for lightning-fast requests.
Day 09: Fixing the Page Reload: Smooth AJAX and Chat Memory
Upgrading our chatbot: adding session memory so it remembers past messages, and switching to smooth AJAX fetch requests so the page never reloads.
Day 10: Demo Day: Presenting My AI Chatbot to the Jury
Final day at Virtual Height! We run final tests, present our capstone AI Chatbot live to the evaluation jury, and defend our architecture in the Viva Q&A.
Weekend 02: Beyond the Code: Communication and the Sales Funnel
The internship wraps up. Reflecting on why commercial training programs push paid upsells, and the undeniable reality that good code must be backed by great communication.
Epilogue: Two Weeks as a 17-Year-Old Intern, and What Comes Next
Closing the 2-week internship chapter. Looking back at what worked, what I learned, and setting my sights on the next wave of independent builds.
Ecosystem Takeaways
Key Learnings
- The massive speed advantage of self-directed project building over classroom lectures.
- Recognizing how corporate training programs function as marketing funnels for paid courses.
- Structuring full-stack projects cleanly (static assets, templates, backend logic).
- Persisting conversation history between messages using Flask session cookies.
- Connecting a Flask backend to remote AI models using the Hugging Face Inference API.
- Starting a local Flask web server and declaring URL endpoints.
- Structuring web page elements cleanly with semantic HTML tags.
- The physical and mental toll of the traditional 9-to-5 corporate loop.
Tools & Stack
- Retrospective
- Observation
- Flask
- Flask Sessions
- HTML5
- Self-Reflection
- Gemini API
- Python 3.11
- VS Code
- Critical Thinking
- Hugging Face API
- Fetch API
Challenges Overcome
- Summarizing two intense weeks of code, notes, and experiences into an honest retrospective.
- Navigating soft sales pitches during evaluation sessions while keeping professional composure.
- Handling live demo pressure and making sure the API responded smoothly on the office network.
- Keeping session cookies secure and handling edge cases where sessions expire.
- Handling API rate limits when sending multiple quick messages.
- Fixing 405 'Method Not Allowed' errors by making sure the route accepts POST requests.
Startup Integration Reflections
In a modern agency ecosystem, AI integration, Python automation, and Prompt Engineering are not peripheral utilities — they form the high-leverage engineering stack that bridges business logic with scalable generative automation. Aligning non-deterministic models using structured prompts and solid Python glue code is the absolute prerequisite for production-grade intelligence.