Freedom Dumlao
1.7K posts
Freedom Dumlao
@APIguy
CTO + CAIO. Ruby Central Board Member. Maker. Builder. Opinions my own.
Boston, MA Katılım Ağustos 2008
693 Takip Edilen1K Takipçiler
English
@zeeg I'm translating this as... them solving 42 second problems in 4.2 seconds... which is still not a problem that I give all that much of a shit about... yet
English
@JasonSwett Carpenters motto:
"Do your best and caulk the rest"
Tewksbury, MA 🇺🇸 English
@JasonSwett I always call things what they are, but literally:
First integer I define? integer_1
Tewksbury, MA 🇺🇸 English
The naming principle that LLMs and human devs hate, hate, HATE to follow: call things what they are.
English
If "software creation is free now" then how come no one has found room in their now-unlimited budget to fix all the bugs that I encounter on a daily basis?
Maybe because "software creation is free now" misunderstands the basic workings of reality and was never remotely true.
English
@rushing_andrei I had this thought right after the Fable takedown. Reminded me of the days when exporting any encryption greater than 40bit was a crime
English
Local models are actually pretty good now.
No, they aren't a reliable source of facts and the baked in knowledge is pretty rough. Done use them for what they "know
Use them for what they can "do"!
They are great for using tools, managing workflows, producing structured outputs, condensing and summarizing text, classification and so much more.
If you install a local model and ask it to recite some facts or tell you about some historical concept it will likely fail. Give it some tools, some memory, and a harness though and WOW
Boston, MA 🇺🇸 English
Anyone who knows me knows I've had an obsession with DSLs for as long as I've been writing code. The idea of a shorthand language for expressing the solution to specific problems is so powerful. And now in the world of AI there is an entirely new reason for DSLs to exist!
Dave Blundin@DaveBlundin
The world is going to be desperately short on compute for the rest of time. No matter how many chips we make, demand will outrun supply. Scaling models in parameter count, context length, and thinking tokens continues to yield increasingly better models with consistently slower, more costly responses. This points to a future with giant frontier models doing some things, but the vast majority of tokens will run through thousands of small, fine-tuned models. Some may even operating with their own domain-specific language. There’s no technical limitation preventing a fine-tuned Gemma 12B model from outperforming Opus 4.8 at a specific task, like linting. But it would run at less than 1% of the cost. Everyone keeps telling young founders, "AI will do everything, there's no purpose in life." Wrong. Domain-specific models and domain-specific languages will be a goldmine in the next few years.
Reading, MA 🇺🇸 English
@cheuk_baby @DaveBlundin It's true! But you have options.
Tewksbury, MA 🇺🇸 English
The leading LLMs are so guardrailed. So, how do you deploy AI into sensitive areas like healthcare, biotech, and financial services?
Bypass the harness and go straight to the model via Bedrock, says Freedom Dumlao (@APIguy). I’ve made the transition myself already...mostly so I can easily log to S3 and keep my completions from Sam/Dario - but this is an added benefit!
English
What you get from the LLM providers is wrapped in guardrails which are limiting what you can do. Doubly true if you access the models via the browser "chat" interface.
Dave Blundin@DaveBlundin
The leading LLMs are so guardrailed. So, how do you deploy AI into sensitive areas like healthcare, biotech, and financial services? Bypass the harness and go straight to the model via Bedrock, says Freedom Dumlao (@APIguy). I’ve made the transition myself already...mostly so I can easily log to S3 and keep my completions from Sam/Dario - but this is an added benefit!
Tewksbury, MA 🇺🇸 English
@DannyOkec Like getting a glimpse of the future only to have it ripped away.
Tewksbury, MA 🇺🇸 English
@APIguy I feel divorced . Just few days in fable and back at other models and it feels like a time travel
English
Another amazing conversation with @DaveBlundin getting into some of the nitty gritty details of making AI accessible to everyone and a lot more. What a blast as always, and thank you for having me on the show! 🚀 💪
Dave Blundin@DaveBlundin
Just posted Round 2 with Freedom Dumlao (@APIguy), CTO & CAIO at @Vestmark, and somehow the conversation got even better than the first one! A few weeks ago, Freedom told us how all 400 Vestmark employees have access to Claude Code. In this episode, learn about the freshly installed rack of Mac Minis now running AI agents at Vestmark, why running inference through AWS Bedrock instead of the Claude.ai API is an unlock for enterprise, and why small fine-tuned models like Gemma 12B are about to eat a huge chunk of the enterprise AI workload. Thank you Freedom for returning to the Link Ventures podcast studio!
Tewksbury, MA 🇺🇸 English
Fine tuning small models is the skill most engineers are still missing from their tool belt. If you aren't learning how to do this yet you are way behind.
Akshay 🚀@akshay_pachaar
Google just dropped a new LLM! You can run it locally on just 8GB RAM. Let's fine-tune this on our own data (100% locally):
Tewksbury, MA 🇺🇸 English
@rushing_andrei @gauravsapkotanp Spaceship == Namecheap
Oak Park, NC 🇺🇸 English
Hey devs,
........ where do you usually buy your domains?
1. GoDaddy
2. Namecheap
3. Cloudflare
4. Porkbun
5. Dynadot
6. Squarespace
7. Hostinger
8. Other....
English
@AlephOneAI @DaveBlundin Well you can tell them you know a guy in business in this economy (and doing very well) and this ☝️is what he says
Oak Park, NC 🇺🇸 English
@DaveBlundin @APIguy It is funny. When AI celebrities say this, everyone agrees.
But when I say something similar as a teacher to my students or to the companies we work with people often tell me I don’t understand how business and the economy really work.
English
If you wait for AI to be "ready," then you're already behind.
Someone who saw it coming is already way ahead of you. The edge goes to whoever prepares now and is ready the moment the door opens.
One of my favorite moments from this AMAZING episode with Freedom Dumlao (@APIguy) from @Vestmark 👇
English
If you're still waiting for AI to be "ready," you are already way way behind.
Dave Blundin@DaveBlundin
If you wait for AI to be "ready," then you're already behind. Someone who saw it coming is already way ahead of you. The edge goes to whoever prepares now and is ready the moment the door opens. One of my favorite moments from this AMAZING episode with Freedom Dumlao (@APIguy) from @Vestmark 👇
Woodfin, NC 🇺🇸 English
By far the most common comment I'm getting from my most recent podcast appearance: "why don't you have any stickers?!"
It's a new laptop people! Which stickers should get first dibs?
English
Had absolute blast talking with the one and only @DaveBlundin - it's always a great conversation and love swapping notes on AI, and the future of work
Dave Blundin@DaveBlundin
It was an absolute pleasure to sit down with Freedom Dumlao (@APIguy), CTO & CAIO at @Vestmark where he and the team are shipping AI inside a highly regulated, $2T+ wealth tech platform. In this episode, learn about how all 400 employees at Vestmark use Claude Code, how they've built agentic systems like Pulse that prompt the advisor instead of waiting to be prompted, and how Fabric allows anyone in the company to spin up and share skills in 3 minutes. Most banks are still rolling out Copilot to "a handful of test users this summer." Freedom is at least 18 months ahead of that and gaining steam. Thank you Freedom for joining me in the @Link_Ventures podcast studio!
Tewksbury, MA 🇺🇸 English