Mistire Daniel

@Mistire37

SWE Student at AASTU | Gen AI Engineer Trainee at @10acad

Addis Ababa, Ethiopia

Joined February 2022

269

Following

Followers

Replies Reposts Media

Mistire Daniel
@Mistire37
May 5
If you're using Qwen3 and stripping `<think>` blocks after generation — you're paying for tokens you never use. And they're slowing down your decoder too. Here's the mechanism.
1
Mistire Daniel
@Mistire37
Apr 18
Check out my latest article: Two Weeks, One Benchmark, Six People: What We Actually Learned Building a Production Data Agent linkedin.com/pulse/two-week… via @LinkedIn
linkedin.com
Two Weeks, One Benchmark, Six People: What We Actually Learned Building a Production Data Agent
Two weeks ago our team started with a challenge: build a production-grade data analytics agent and compete on UC Berkeley's DataAgentBench a benchmark designed to expose exactly why AI data agents...
Mistire Daniel
@Mistire37
Apr 18
Two weeks ago we started building a data agent from scratch. Today we're submitting to UC Berkeley's DataAgentBench. Here's what we actually built and what we learned the hard way.
1
Mistire Daniel
@Mistire37
Apr 14
Most AI demos show a clean question going in and a clean answer coming out. What they don't show: the same customer stored as integer 10023 in one database and 'CUST-10023' in another. We hit this on Day 1 building a data agent for real enterprise data.
1
Mistire Daniel
@Mistire37
Mar 16
Week 4 at @10acad: built a “Brownfield Cartographer” to analyze unfamiliar codebases. Tried adding a UI, API, and Docker, which made debugging much harder and took a lot of time. Also hit LLM API quota limits, so some features didn’t fully run. #10Academy #My10AcademyTRP1Week4