Oct 16, 2019 · 34m · saastr
Founder's Guide to Scaling Applications: When to Build, When to Buy and What Breaks
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
Algolia co-founder and CTO Julian Lemoine shares critical lessons on architectural scaling, explaining why engineering teams must resist building non-core tools internally and instead purchase third-party solutions unless custom development provides a 10x competitive advantage.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. How this is scored →
Jason as informed peer 0.0 Guest teaching 0.0 Guest disagreement 1.1 Jason pushing back 0.0
speaking balance: gold is Jason, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 21:50 Internal teams cannot compete with SaaS
Julian firmly dismisses the illusion of internal tool development, asserting that in-house teams can never outpace a focused external SaaS company over time.
Hardest push from Jason ▶ 1:55 Monologue presentationBecause this episode is a solo keynote without conversational host dialogue, no host pushback occurred.
Biggest teaching moment ▶ 27:00 The real cost of engineering timeJulian delivers a masterclass on total cost of ownership, explaining that burning senior engineering hours to avoid expensive SaaS vendors is a massive net loss.
Jason holds their own ▶ 31:32 Solo keynote deliveryThe presentation format contains no host interventions or pushback exchanges.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Jason as informed peer | Guest teaching | Guest disagreement | Jason pushing back | Why |
|---|---|---|---|---|---|---|
| The Engineering Paradox in Build vs Buy Decisions | 0 | 0 | 1 | 0 | Julian delivers a solo presentation introducing the paradoxical downsides of having strong engineering teams who instinctively want to build rather than buy. Because this is a monologue presentation without active host participation, host-side metrics are scored at zero. | |
| Case 1: Early-Stage Bare-Metal Infrastructure Bet | 0 | 0 | 1 | 0 | Julian outlines Algolia's early contrarian bet on bare-metal servers to deliver a 10x performance differentiator. The presentation is purely informative and reflective, outlining high-risk architecture choices in early startup stages. | |
| Case 2: Distributed Search Network and DNS Architecture | 0 | 0 | 1 | 0 | Julian explains the dilemma of building a custom DNS engine versus finding a specialized early-stage startup vendor. The segment is an instructive monologue highlighting how engineers easily underestimate long-term edge-case complexity such as DDoS protection. | |
| Case 3: Infrastructure Monitoring and Open-Source Prototypes | 0 | 0 | 2 | 0 | Julian breaks down the internal friction of managing engineers who built a Grafana/Graphite monitoring prototype in a single day. He advises engineering leaders to challenge assumptions on operational costs rather than issuing top-down mandates. | |
| Case 4: Search Analytics, Scaling Limitations, and Maintenance Debt | 0 | 0 | 1 | 0 | Julian walks through Algolia's iterations on analytics infrastructure and the trap of spending elite engineering hours maintaining crumbling internal systems instead of temporarily paying expensive SaaS pricing. | |
| Framework for Build vs Buy: Factor 10 and Technical Debt | 0 | 0 | 1 | 0 | Julian synthesizes his build-vs-buy framework, stressing that internal development should be reserved strictly for core 10x differentiators to avoid decades of legacy technical debt. | |
| Audience Q&A: Infrastructure, Vendor Selection, and 10x Differentiation | 0 | 0 | 1 | 0 | Julian answers audience questions regarding infrastructure ownership and vendor evaluation, emphasizing roadmap alignment over feature-list comparisons. |
Statements from this episode (13)
Algolia ran 3,000 bare-metal servers handling 300B API calls in 2019
“We are today around 350 people in six different offices. We have mainly a bare metal infrastructure with close to 3000 servers distributed in more than 70 data centers, and we handle between 203 hundred billions API columns.”
Julian Lemoine: Strong engineering teams struggle more with build-vs-buy decisions
“The better your team is, the more difficult it is to take the good build versus by solution.”
Algolia founders chose bare-metal hardware over cloud VMs at inception
“First big decision we took, it was in the very early days of the company, we were only two, we have not even fundraise, we have no salaries, so very, very early days of the company, we decided to use bare metal infrastructure and not use cloud infrastructure, …”
Cloud infrastructure in 2012 lacked the hardware required for search engines
“Like search engines are very intensive in terms of CPU, SSD, memory, so we had to pretty much select a very specific machine, very specific hardware, to make sure we have the best performance of the market. None of that was available on cloud infrastructure at…”
AWS Route 53 supported only 7 of Algolia's 12 regions in 2014
“Like we had at the moment of the launch, 12 regions. Amazon was covering only seven of them. So we were not able to do the geo routing for all our customers, all our region.”
Julian Lemoine: Engineers overconfidently propose building tools outside their core expertise
“I think a good engineering team will always find this, and they will come to you and give you a big list of things you could do way better than anyone else, even if they have no strong expertise on this market.”
Julian Lemoine: Challenge engineers on maintenance costs instead of arguing
“So what I did learn in this story is never try to convince them, but challenge them on the cost of developing their prototype, challenge them on the fact we cannot use our existing monitoring solution for six months to a year while we find a good solution on t…”
Julian Lemoine: Internal engineering teams can never compete with SaaS vendors
“And I think that the big learning with SAS, your internal team will never be able to compete with the SAS product. Never. It's hopeless. Like, if the solution is successful, the team will scale, and maybe today they have two engineers, but in two years they wi…”
Julian Lemoine: Startups prematurely cut tooling costs instead of buying engineer time
“So that's another mistake, like looking at the cost, maybe too early, and spending some money for a limited period of time would basically buy us some time, and the time of one of our best engineers, which is a big thing, like we never have too many engineers.”
Julian Lemoine: Design replaceable components instead of permanent internal systems
“One of the best ways to deal with it is to accept we can replace components easily over time, and don't think about building something for the long term.”
Julian Lemoine: 99% of startups should not build their own infrastructure
“I think 99% of the company, or 99%, should not build the infrastructure. I think you need to have a big bet, and it needs to be a huge difference in your product, in your company, in your business, to have a good reason to build it.”
Julian Lemoine: Evaluate SaaS vendors by their roadmap, not current features
“I never want my team to think in terms of features. I think you have the product with the feature set as it is today, but if you use and you buy a solution, you need to project your decision on the long term. So looking at the roadmap, discussing the roadmap w…”
Julian Lemoine: Only build in-house if it delivers 10x competitive differentiation
“I think there is only one for me, which is a factor 10. Like, if you have this factor 10 on one specific area which is important for your business, it can be the UX, it can be the performance, it can be the relevance, it can be anything. If you have this facto…”