AI Architect | Writer, thinker, and lifelong learner. Subscribe to my newsletter: https://xitter.twitterapi.xyz/t.co/1uuir2x6fZ
Bengaluru, India
Joined September 2018
- Tweets2.6K
- Following80
- Followers4.7K
- Likes1.6K
Averaging your LLM judges gives the weakest one the same vote as the best. On a 200-item gold set, you can’t tell them apart.
AGORA (arXiv, Nov 3) uses item-response theory to learn each judge’s reliability without labels.
How to try it: arxiv.org/abs/2611.05556
2
596
Charging platform fee by e-comm by platform is like if a local shopkeeper ask customers a portion of rent for the shop
178
Coding agents are very good at building what you ask for.
The gap between what you specify and what you expect is where the agent starts making assumptions.
More assumptions → more rework → more iterations → more tokens.
The clearer the specification, the closer the output.
🤖 Made with AI
1
163
Many people want AI to handle their thinking, execution, and everything in between—which surrenders their leverage. I think, AI should execute and retrieve information for me, but I keep the thinking.
1
181
Many times users aren’t clear on their requirements for an LLM. They expect surprisingly good results even when they themselves don’t know what output they’re looking for.
1
2
282
Is it just me, or does Codex hit its usage limits much faster than Claude for the exact same task under identical conditions?
1
221
Two agents can use the exact same LLM and still be radically different systems.
Give one:
full filesystem + network + shell
Give another:
sandbox + scoped files + restricted tools
Same intelligence. Different capability. Different blast radius.
Agent = Model + Harness
2
2
469
The interesting part of “agentic video” isn’t that an AI can watch a 90-minute video.
It’s that it doesn’t have to.
Instead of processing every frame equally:
Question → search → candidate moments → inspect deeply → answer.
AI is starting to reason about computation.
1
208
Romee Panchal retweeted
A useful rule in AI engineering:
Don’t start with the model. Start with the failure mode.
Ask:
• What can go wrong?
• Can rules solve part of it?
• Where does ML add value?
• How will we know it works?
Good AI starts with good problem framing.
Why do Swiggy's Bolt orders almost always take as long or longer than regular orders? I have never seen them deliver on time.
These people need to work on their estimation algorithms.
@Swiggy @SupportSwi88934
2
206
Telecom companies continue pushing their higher data plans to boost their revenues.
I like that 1.5GB/day annual plan. However, its no longer available.
1
224
Self-doubt hits even when you’re doing meaningful work.
Mistakes feel bigger than they are. Comparison makes you question your progress. AI tools can help, but they can also weaken trust in your own thinking if you overuse them.
romeepanchal.com/posts/life/…
1
247
Most AI systems don’t fail because the model is bad.
They fail because the system around the model is under-designed.
Read the agentic engineering design patterns here:
romeepanchal.com/posts/ai_en…
1
207
Six Pattern Families Every AI Systems Engineer Needs to Know romeepanchal.com/posts/ai_en… #AIengineering #LLMsystems #productionAI #systemdesign