Eric Feng

sqlc is nice. I’m never going back to ORMs. It feels right that Postgres interfaces are defined in .sql and not a DSL. One of the unexpected benefits: all my DML is “type checked” against the schema, so if one of my queries references a deleted column, sqlc refuses to compile.

Notes from the cold emails I’ve both sent and received.

I’ve been struggling to make sense of open source since college. I remember installing Linux for the first time and marveling that I could get an operating system for free. Why are people giving away something so valuable?

There should be no such thing as free public parking.

In the 🔗 xi editor retrospective, Raph touches on modularization’s costs that I have experienced myself.

Enterprise sharing. When Numeracy was acquired by Snowflake, sharing was one of the core features they wanted to integrate. One year later

This post was prompted by Ben Thompson’s discussion on the latest open source angst. His post is not the first, nor is it the more comprehensive, but it best captures the heart of the issue for me.

I recently calculated trial conversion in Stripe, but found it remarkably hard to find historical trials. In case others have need, here’s what I found.

I created Faker.sql, my first Postgres extension for generating mock data.

Focus.

How long is a customer waiting for reply?

Tried out ics.py and icalendar to parse Google Calendar calendars with Python (without having to write a custom integration).

I’ve been setting up a Sigma-like replica of Stripe. However, unlike the Gmail API, the Stripe email isn’t built for data sync, but for…well, billing flows.

Pricing decisions are multi-faceted (your product has an infinite number of potential pricing scheme and price points) and each part of the decision is susceptible to bike shedding.

Following up on my last post and a Twitter exchange and of course, musing on commodity pricing, I decided to play around with Google Cloud Functions to see if it was a viable alternative to tools like Segment, Mixpanel.

This year, I’m going to start cold emailing customers.

It’s best to define a job by the projects and their results. Projects vary by company, but results are generalizable across businesses (especially SaaS).

Flask is my favorite server framework, due to its clarity and flexibility. However, getting Flask beyond “Hello, World” always feels iffy, with most popular online tutorials being several years old and having been built by educators instead of veterans.

A predominantely self-taught application developer learns to deploy with Google Container Engine.

If a tree falls in the forest and no one is around to hear it, does it make a sound?

Disclaimer: I know very little to nothing about how investors value private companies. But I’ve never read an article that even attempts to do basic arithmetic around this, so I thought I’d write down my own thoughts.

To those hooking into the internals of Jekyll, knowing a bit more about what each variable includes can be helpful.

You see growth engineering in consumer tech (Facebook, Pinterest, Uber), but I’ve seen less of it for B2B SaaS, so one year in, I wanted to share my understanding of growth engineering as it fits into our business.

Taking note on why I live.

Working in a startup, it’s easy to forget all the random things I’ve done. A brief retrospective.

I’ve found Gmail’s API surprisingly confusing, mostly because email was so clearly built in a time before JSON. The message payload is broken into parts, depending on the structure (attachments, images) of the email, each labeled with a different, poorly documented mimeType.

We’ve been using Redash at Sentry lately. I’d never heard of it before (Mode being the frontrunner in this category), but it’s damned good so far. And before you ask me about Metabase, it’s a totally different use case.

Itinerary: Tokyo and KyotoMy Map

I want it to rain.