Elizabeth Tran
Elizabeth Tran

Builder &
operator

What I do

I talk to people, find the real problem, and build what fixes it, utilizing AI.

See my work ↓

Babson College ’28
Business · Technology Entrepreneurship

2M+

TikTok views on content I made

$100K

Solutions supported at AI Technology Partners

3,000+

Students reached

80+

Small businesses consulted

01 / Work

Selected work

01

Daily Cash Agent

AI Agent · Plaid · Fintech

Shipped
Sample data

dailycashbot

bot

Operating account

Current balance
$12,480.00
Pending charges
$1,215.40
Net tomorrow
$11,264.60

6:45 AM

Problem
A founder checked 6 different bank accounts by hand every morning to get one number.
What I learned
He didn't want a dashboard. He wanted the answer pushed to him.
What I decided
One Telegram message (balance, pending charges, net cash) instead of an app.
Result
Runs unattended every morning. The client still uses it and recommends it.

02

Babson's Largest AI Hackathon

Partnerships · Community

Shipped
Hackathon participants with The Generator banner outside Richard Knight Auditorium, Babson College
Problem
The Generator needed industry weight behind its flagship build event.
Approach
Warm intro outreach to land sponsors, plus direct outreach to schools around Boston.
Result
3 sponsors (Anthropic, GitHub, and Cursor) and 500+ students from 50 schools at a full-day AI hackathon.

03

AI Sales Agents

AI Agents · n8n · B2B

Shipped
Problem
B2B pipelines lose deals in the gap after a discovery call.
What I built
Agents that qualify leads, draft personalized follow-ups, and hold context across client conversations.
Result
Post-call follow-up and stakeholder tracking run without a human in the loop.

04

Newsletter AI Automation

AI Agents · Claude Code · Content

Shipped
Problem
C-suite readers needed a briefing nobody had hours to research.
What I built
Agents that monitor sources, extract key insights, and draft executive-ready summaries.
Result
Hours of manual research a week, removed.

05

Small Business AI Bootcamp (G1000 Program)

AI Consulting · API Keys · MCPs

Ongoing
Problem
Owners with real operational problems and no way into AI tooling.
What I did
Live demos, API key setup, and MCP integrations, from automating customer responses to marketing workflows.
Result
80+ small business owners advised on putting AI into daily operations.

02 / Testimonials

In their words

The people I built for.

Client · Daily Cash Agent

Liz did an outstanding job building an AI-powered Daily Cash Management agent that has become an incredibly valuable tool for me. She continues to provide excellent support, proactively maintaining the agent and quickly resolving any bugs that arise, making the entire solution reliable and hugely helpful to our daily workflow. I would recommend her without hesitation.

David
DavidCo-founder, Metal Fuels · Boston, MA

Student · AI Agents Course

The course was amazing and opened my mind to different paths that I could venture too. Additionally, the lessons were very helpful and straightforward, which is nice for someone who is a slow learner

5 sessions · 10 hours · 1:1

Agent fundamentals, Claude Code, and building and selling his first AI agent.

Brandon
BrandonHigh school student · AI Agents Course

03 / About

About

I'm a junior at Babson studying Business with a concentration in Technology Entrepreneurship. I'm Partnerships Lead at The Generator, Babson's AI lab, and Student Lead for the AI & Small Business Bootcamp (G1000 Program).

Building custom AI workflows for two paying clients ($3,000+ in revenue) taught me the most important lesson fast: most clients ask for a tool, but what they need is a clearer problem.

How I work

  • Talk to users before building

    I ask what someone does every morning before I ask what they want built.

  • Scope the smallest useful thing

    One Telegram message beat a dashboard for the Daily Cash Agent.

  • Write it down

    One-pagers and clear plans, so we disagree on paper instead of in production.

  • Stay technical enough to prototype it myself

    Claude Code, n8n, and APIs mean an idea gets tested this week, not next quarter.

Let's build something