Skip to main content
The Changelog

From GitLab to Kilo Code (Interview)

77 min episode · 2 min read
·

Episode

77 min

Read time

2 min

AI-Generated Summary

Key Takeaways

  • Parallel Treatment Strategy: When doctors declared standard care exhausted, Sijbrandij combined multiple experimental treatments simultaneously rather than sequentially, prioritizing cure over research reproducibility. He assembled a tumor board including pathologists to ensure safe drug combinations that don't overload organs like kidneys.
  • FAP-Targeted Radioactive Therapy: Single-cell sequencing revealed high fibroblast content in his tumor, enabling targeted treatment in Germany using FAP binders combined with lutetium radioactive elements. This achieved 60% necrosis and 20% shrinkage with zero detectable side effects, detaching tumor from spinal cord to enable surgical removal.
  • Maximal Diagnostics Approach: Fresh-freezing tumor samples instead of standard FFPE preservation enables running hundreds of additional diagnostic tests later. Sijbrandij traveled to China for first-available B7-H3 scans, discovering highest expression levels seen in 20 patients, which directly informed drug development modifications to protect liver function.
  • Kilo's Multi-Model Architecture: The platform supports over 500 models including eight of ten most popular being free to use, charging exact list price for paid models without markup. Users create profiles combining specific models with custom prompts, enabling teams to share optimized agents for tasks like Java upgrades.
  • Agent Productivity Multiplier: Kilo engineers ship at speeds equivalent to seven-to-eight person teams from two years ago by parallelizing work across multiple AI agents. One engineer can manage multiple agents working simultaneously on separate tasks with small context windows, reducing costs while increasing speed and preventing context loss.

What It Covers

Sid Sijbrandij, GitLab founder, shares his bone cancer journey since 2022, including experimental treatments across three continents, single-cell sequencing diagnostics, and launching six biotech companies plus Kilo, an open-source all-in-one agentic coding platform.

Key Questions Answered

  • Parallel Treatment Strategy: When doctors declared standard care exhausted, Sijbrandij combined multiple experimental treatments simultaneously rather than sequentially, prioritizing cure over research reproducibility. He assembled a tumor board including pathologists to ensure safe drug combinations that don't overload organs like kidneys.
  • FAP-Targeted Radioactive Therapy: Single-cell sequencing revealed high fibroblast content in his tumor, enabling targeted treatment in Germany using FAP binders combined with lutetium radioactive elements. This achieved 60% necrosis and 20% shrinkage with zero detectable side effects, detaching tumor from spinal cord to enable surgical removal.
  • Maximal Diagnostics Approach: Fresh-freezing tumor samples instead of standard FFPE preservation enables running hundreds of additional diagnostic tests later. Sijbrandij traveled to China for first-available B7-H3 scans, discovering highest expression levels seen in 20 patients, which directly informed drug development modifications to protect liver function.
  • Kilo's Multi-Model Architecture: The platform supports over 500 models including eight of ten most popular being free to use, charging exact list price for paid models without markup. Users create profiles combining specific models with custom prompts, enabling teams to share optimized agents for tasks like Java upgrades.
  • Agent Productivity Multiplier: Kilo engineers ship at speeds equivalent to seven-to-eight person teams from two years ago by parallelizing work across multiple AI agents. One engineer can manage multiple agents working simultaneously on separate tasks with small context windows, reducing costs while increasing speed and preventing context loss.

Notable Moment

Sijbrandij received a text saying his PET scan was positive not subtle, revealing 50 metastatic lung sites that appeared terminal. After telling his team and receiving condolences from multiple doctors, the final specialist recognized it as COVID remnants, not cancer, based on lymph node patterns.

Know someone who'd find this useful?

You just read a 3-minute summary of a 74-minute episode.

Get The Changelog summarized like this every Monday — plus up to 2 more podcasts, free.

Pick Your Podcasts — Free

Keep Reading

More from The Changelog

We summarize every new episode. Want them in your inbox?

Similar Episodes

Related episodes from other podcasts

This podcast is featured in Best Cybersecurity Podcasts (2026) — ranked and reviewed with AI summaries.

You're clearly into The Changelog.

Every Monday, we deliver AI summaries of the latest episodes from The Changelog and 192+ other podcasts. Free for up to 3 shows.

Start My Monday Digest

No credit card · Unsubscribe anytime