Data Scientist · AI Engineer · Artist
I build systems
that explain themselves.
Data Science Apprentice on the 718 Digital Labs team at The Knowledge House — building pipelines, applied AI agents, and LLM features on live projects. Off the clock I paint, sculpt, and draw, which is really the same job: take something tangled and make it legible to somebody else.
How I work
I learn by doing, and I keep honing a craft to keep it. Complex concepts are the fun part — I want to point them at a real problem the same week I learn them. Feedback is how I find out what actually matters to people, so tell me both halves: what’s landing and what isn’t.
Data & analytics
AI & LLM engineering
Build & ship
Section 02
Tech Projects
ALVI — RAG pet-care assistant
ALVI translates the data your pet creates every day into language you can understand. Custom scraper feeding a vector knowledge base, a full RAG pipeline from semantic chunking through context injection, and a system-prompt architecture that governs scope and safety.
What the apprenticeship changed
Real code review at 718 Digital Labs reshaped ALVI’s structure — evaluation logging before feature work, and schema discipline on the knowledge base.
NYC School Performance Analysis
A multi-source Python pipeline joining NYC OpenData school metrics with MTA transit data, then KMeans clustering to surface how transit access tracks with student outcomes. Published as a Tableau portfolio.
Day job · 718 Digital Labs
Pipelines and Python analysis answering live product questions; LLM integrations, prompt-engineering experiments, and applied AI agent work — all in Agile with engineers and product partners.
Section 03
Credentials
Training completed and work signed off by the people who received it.
Innovation Fellowship — Data Science
The Knowledge House
Certificate of completion for the 9-month 2025 Innovation Fellowship on the Data Science track.
Signed by
Jerelyn Rodriguez, Co-Founder & CEO
Joe Carrano, Co-Founder & Strategic Advisor
“Create Standardized Measures to Assess Student Outcomes”
Prometheus · Pro Bono Project Completion
Delivered a Python, SQL, and Google Sheets solution to a nonprofit working in youth development, education, and mental health.
Confirmed by
Margaret Smiley, Executive Director
Section 04
Testimonials
What people say after working with me directly.
“
Thalyann leveraged SQL and Python to deliver a program evaluation to an education nonprofit, using these skills to create standardized measures and assess student outcomes… Through…
Margaret Smiley
Executive Director, Prometheus
“
I had the pleasure of working with Thalyann on a data analytics project for a nonprofit through Prometheus. As team lead, she balanced technical expertise with clear communication,…
Mei H.
Data Science Teammate, Prometheus
Section 05
Lab Notebook
Short build-log entries: what I tried, what broke, what I changed. Failure modes are the most useful thing I own — this is where they live in public.
Month three: learning enough to ship
Two codebases, eight tickets, and one week spent trying to understand everything before touching anything — what month three of the 718 Digital Lab apprenticeship taught me about contributing before I feel ready.
Proof of (Artificial) Life: Getting Nova Micro to Talk Back
A five-year-old Coursera course, a Bedrock model that really wanted to talk about dance, and one AWS Skill Builder lab that was already out of date.
The build catalog: three things shipping before year-end
ALVI's beta, a verified Enigma machine on an ATtiny85, and finishing the NYC schools × transit analysis — what's on deck before December.
Section 06 · Let’s talk