Hacker Newsnew | past | comments | ask | show | jobs | submit | gbrits's commentslogin

Location: Netherlands / GMT+2 Remote: Yes. Experienced with UK and US clients; open to timezone overlap. Willing to relocate: No. Occasional travel is fine.

Technologies: Node.js/TypeScript, ClickHouse, PostgreSQL, Redis, Elasticsearch/OpenSearch, DuckDB, Vue, Tailwind, event-driven systems, low-latency data pipelines, analytics infrastructure, and hybrid search/retrieval. Also comfortable with Python and Rust.

Résumé/CV: Available upon request LinkedIn: https://www.linkedin.com/in/geert-brits/ Email: gbrits<at>gmail<dot>com

Hands-on Staff/Principal engineer and architect with 20+ years of experience building backend systems, data-intensive products, search and retrieval infrastructure, internal tools, and applied-AI workflows.

I’m most useful when the business goal is understood but the product and technical shape are not. I help clarify what should be built, make tradeoffs explicit, shape the architecture, and build the critical parts myself. Backend is where I’m most at home, but I regularly work across APIs, data, frontend applications, analytics interfaces, and production operations. I bring strong product and UX judgment without moving away from implementation.

Selected work:

- Principal engineer and architect on a ground-up KPI and analytics platform rewrite with a five-person team, including its data models, self-service dashboards, and drill-down workflows. The company later exited for approximately €10M.

- Technical founder of a market-intelligence SaaS matching more than one million dynamic segments against continuous on-chain and off-chain streams, backed by ClickHouse pipelines processing billions of rows per day.

- Rebuilt a live B2B marketplace using Node.js, Vue, and Elasticsearch, including saved-search matching and notification infrastructure. Currently improving its search quality using hybrid retrieval, sparse-vector signals, reciprocal rank fusion, and reranking.

I prefer freelance and contract work as a hands-on senior IC at Staff or Principal level. Founding Engineer roles are also a strong fit. I’m open to full-time opportunities when the product, team, and role are an exceptional match.

Best fit: backend-heavy products, applied AI and workflow automation, analytics infrastructure, search and retrieval systems, complex internal tools, or early-stage teams that need architecture, product judgment, and implementation depth. My preferred leadership track is through product shaping, technical direction, mentoring, and hands-on engineering rather than people management.


Indeed. Or shorter: “auto”

Location: Netherlands / GMT+2

Remote: Yes. Experienced with US / UK clients; open to timezone overlap.

Willing to relocate: No, occasional travel OK.

Technologies: Backend architecture, data-intensive product systems, event-driven systems, low-latency pipelines, analytics infrastructure, search/retrieval, AI, Node.js/TypeScript, ClickHouse, DuckDB, PostgreSQL, Redis, Elasticsearch/OpenSearch, Vue, Tailwind; also comfortable with Python and Rust, and quick to become productive in unfamiliar stacks.

Résumé/CV: Available upon request

LinkedIn: https://www.linkedin.com/in/geert-brits/

Email: gbrits<at>gmail<dot>com

Hands-on principal engineer with 20+ years across backend systems, data-heavy products, search/retrieval, internal tools, and founder-led product work. Technical co-founder, CTO/CPO, architect, fractional technical lead, and engineering owner.

I work best where the business goal is clear but the technical shape still needs senior ownership: unclear requirements, fragmented tools and data, API boundaries, retrieval quality, and architecture that has to stay understandable as the product grows. I can own the technical architecture and system shape where needed, and I stay close enough to implementation to keep latency, operability, failure modes, and tradeoffs grounded in reality.

I work across the whole product system: internal tools, backend services, data/retrieval layers, API contracts, observability, and production behavior.

Selected work: Fractional principal engineer / architect for ground-up KPI/analytics platform rewrite (5-person team); company later exited for ~€10M. Technical founder / CTO / CPO for market-intelligence SaaS: live matching of 1M+ dynamic segments against continuous on-chain/off-chain streams; ClickHouse-backed pipelines processing billions of rows/day with near-real-time downstream availability. Marketplace search/matching rewrite (Elasticsearch, percolator-style saved searches, high-volume notification infrastructure) while keeping the live platform in production.

Looking for freelance, contract, fractional CTO, principal engineer, or technical lead roles where one senior person needs to hold the whole system in their head. Also open to full-time opportunities for an exceptional fit. Best fit: backend-heavy products, analytics/data infrastructure, complex internal tools, search/retrieval systems, or early-stage teams that need architecture + implementation depth without a large engineering org.


Location: Netherlands / GMT+2 / Remote: Yes. Experienced with US / UK clients; open to timezone overlap.

Willing to relocate: No, occasional travel OK.

Technologies: Backend architecture, data-intensive product systems, event-driven systems, low-latency pipelines, analytics infrastructure, search/retrieval, AI, Node.js/TypeScript, ClickHouse, DuckDB, PostgreSQL, Redis, Elasticsearch/OpenSearch, Vue, Tailwind.

Résumé/CV: Available upon request LinkedIn: https://www.linkedin.com/in/geert-brits/ Email: gbrits<at>gmail<dot>com

Hands-on principal engineer with 20+ years across backend systems, data-heavy products, search/retrieval, internal tools, and founder-led product work. Technical co-founder, CTO/CPO, architect, fractional technical lead, and engineering owner.

I work best where the business goal is clear but the technical shape still needs senior ownership: unclear requirements, fragmented tools and data, API boundaries, retrieval quality, and architecture that has to stay understandable as the product grows. I can own the technical architecture and system shape where needed, and I stay close enough to implementation to keep latency, operability, failure modes, and tradeoffs grounded in reality.

I work across the whole product system: internal tools, backend services, data/retrieval layers, API contracts, observability, and production behavior.

Selected work: Fractional principal engineer / architect for ground-up KPI/analytics platform rewrite (5-person team); company later exited for ~€10M. Technical founder / CTO / CPO for market-intelligence SaaS: live matching of 1M+ dynamic segments against continuous on-chain/off-chain streams; ClickHouse-backed pipelines processing billions of rows/day with near-real-time downstream availability. Marketplace search/matching rewrite (Elasticsearch, percolator-style saved searches, high-volume notification infrastructure) while keeping the live platform in production.

Looking for freelance, contract, fractional CTO, principal engineer, or technical lead roles where one senior person needs to hold the whole system in their head. Also open to full-time opportunities for an exceptional fit. Best fit: backend-heavy products, analytics/data infrastructure, complex internal tools, search/retrieval systems, or early-stage teams that need architecture + implementation depth without a large engineering org.


Location: Netherlands / GMT+2

Remote: Yes. Experienced with US / UK clients; open to timezone overlap.

Willing to relocate: No, occasional travel OK.

Technologies: Data engineering, backend architecture, event-driven systems, low-latency pipelines, analytics infrastructure, product engineering, Node.js, ClickHouse, DuckDB, PostgreSQL, Redis, SQLite, Vue, Tailwind

Résumé/CV: Available upon request

Email: gbrits@gmail.com

Hands-on CTO / principal engineer with 20+ years in software engineering. Technical co-founder, CTO/CPO, architect, fractional product owner, and engineering lead.

I work best at the intersection of architecture and implementation: designing data-heavy systems well, then staying close enough to the engineering to keep latency, operability, failure modes, and tradeoffs grounded in reality.

Selected work: fractional lead/architect for ground-up KPI platform rewrite; 5-person FTE team; company later exited for €10M. CTO/CPO for crypto/data product; live matching of 1M+ complex dynamic slices against continuous on-chain/off-chain metric streams; high-throughput, near-real-time latency suitable for trading workflows. ClickHouse-backed worker pipelines processing billions of rows/day while preserving near-real-time downstream availability.

Looking for freelance, contract, fractional CTO, principal engineer, technical lead, or data engineering work. Best fit: backend-heavy products, analytics/data infrastructure, complex internal tools, low-latency pipelines, or early-stage teams that need one senior person with both breadth and depth.


Indeed. Extending idempotency to the response not merely the action wherever possible


But why? Now the client has strictly less information than it had before.

The client already has the capability to handle 404 for DELETEs.


Fwiw at least xtc and coke can be legally tested in the Netherlands. https://www.drugs-test.nl/en/testlocations/


They can be in more countries than just the Netherlands IIRC. And simple test-kits are available too - these aren't as good obviously, but can e.g. prove the absence or presence of certain (by)products.

Also, these tests are set up in a way that they are anonymous in a way that e.g. even if prosecutors could subpoena a test-location or lab to hand over administration, this administration won't contain any PII or even pseudonymous information.


I was under the impression that it's impossible to reliably test, e.g., cocaine for fentanyl, without destroying the drug and it doesn't take much of the latter narcotic to kill you.


The amount that has to be destroyed is miniscule.


my point is that powders don't mix evenly. part of a baggie testing negative for fentanyl doesn't mean the remainder is clean.


Then thoroughly mix the whole amount before taking a sample.


Given that a lethal dose of fentanyl is about two milligrams (similar to 5-7 grains of table salt in volume), that still feels pretty risky.


Ah funny! You were 3 hours before me. It's not just xtc and coke. It's much more than that. Most things are simply going to be sent to the lab.

The only thing they don't test are things you can somehow get legally, if I recall correctly (so whatever you can get for any psychological condition).


Lsd can be tested as well in the Netherlands. You can get a quantative analysis, but you lose the sample.


Imagine iPhones with Geiger counters. Nicely aggregated in bite sized daily stats through Apple health


Geiger counters don’t work on Ni-63. It’s a pain to detect, needing liquid scintillation.


Congrats with the launch. This looks great. Inferring schemas on the fly is awesome to get started quickly, but are there ways to explicitly define a schema if I wanted to? For example, thinking of setting column specific compression


Currently no, but I’m open to the idea of being able to set this. Perhaps the ability to set it on the fly (or better yet, inform the user of which type of compression to use based on their actual data!) would be useful. Happy to discuss in a gh issue too.


Yep. Lora's are the easiest way to go. Loads of tutorials on Youtube. This is a good one: https://www.youtube.com/watch?v=70H03cv57-o


Looks like it requires good hardware to run this? My GPU is too old for this.


You typically want an Nvidia GPU with at least 8GB of VRAM to get started with this stuff. You can get away with less, but it will be slow going. More VRAM is better.

If you're serious about learning how to use these tools, it's far more affordable to rent a GPU in the cloud. Google Colab even has some free tiers with limited access to GPUs that are significantly more powerful than what you would normally put in a desktop.


Thanks for the advice, I’ll look into the option of renting a gpu sadly.

I just want to create my own images like Dall-e 2 but using my face instead.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: