The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
Built a reusable Python workflow for cleaning and validating messy Excel and CSV data. It normalizes categories, dates, and text, removes duplicates, flags invalid records, and exports cleaned files with audit summaries.
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Related posts
DataFlow is a self-initiated Python automation tool designed to clean, analyze, and transform messy Excel and CSV files into structured, usable reports.
It detects duplicates, missing values, empty rows and columns, and common formatting issues while preserving the original source file. The project includes a Windows desktop application, automated Excel report generation, and an interactive browser-based CSV cleaning demo.
Built a LangChain-based agentic application that turns a learning goal into a structured, personalised learning plan.
The application supports two execution modes:
⚡ ReAct Agent Mode
A tool-using agent dynamically selects custom tools for course lookup, completion-time calculation, and learning-plan generation.
🔍 Explain Mode — LCEL
A deterministic LCEL pipeline executes the planning workflow step-by-step, making the process more transparent, predictable, and easier to debug.
The project also uses Pydantic structured outputs, Groq, Python, Streamlit, and custom LangChain tools, with deterministic logic kept outside the LLM wherever possible.
This project was built to explore the practical engineering side of AI agents — tool calling, agent orchestration, structured outputs, deterministic workflows, and explainability.
The project is not currently deployed, but the complete source code and implementation are available on GitHub.
Tech: Python · LangChain · ReAct · LCEL · Groq · Pydantic · Streamlit
turning a learning goal into an actual plan is such a useful agent use case 🔥
On a phone call, silence is the bug.
A text chatbot can take three seconds to reply and nobody minds. A voice agent can't. Leave a caller in silence for a beat too long and they say "hello?", start talking over it, or hang up.
That one fact shaped everything we built on Talk-Lee, an AI voice agent that answers business calls for healthcare, real estate and finance teams. Scheduling, support, lead qualification, around the clock.
The goal was a reply in under 500ms. You don't get there with a faster model. You get there by making sure nothing waits for anything else.
→ Speech to text streams while the caller is still talking
→ An LLM and intent layer keeps track of what they actually want
→ Text to speech streams back, so the agent starts talking before the whole answer is ready
→ An orchestrator decides in real time whether to answer, book or hand off to a person
Then it has to do something useful. It books into Calendly, logs the lead in HubSpot, and passes the hard calls to a human with the context already attached.
Where it landed. Under 500ms responses, 1,000+ concurrent calls, 30+ languages, GDPR and TCPA compliant.
Third project in a row with the same lesson. The model is the easy part. The plumbing around it decides whether anyone keeps using it.
