sajal

want to work on the intersection of ai, mathematics, consciousness, dna, memory, hardware

Continuity (continuous functions) is the best language for the world we experience - therefore the foundation behind laws of physics, calculus, neural networks. Not necessarily the truth, but our most powerful approximation; universal approximation theorem.

you rock my world @hkproj ! what a beautiful explanation of HNSW algorithm - youtube.com/watch?v=rhZgXN…

@vanshbulchandan night also

@sajalshlan Locking in for the day is more important

me and claude locking in for the day

going from absolute measures to relative measures in anything is a form of greater intelligence i think

@NirbhikJangid reading, i guess stanford online courses on yt might be a good start

@sajalshlan like joining a college or reading articles/papers?

i want to study economics

Just started reading about Benchmarks and Evals for OCRs, LLMs, VLMs. Such a impactful space to be working at right now: - almost all benchmarks have ground truth errors - integrity of test eval sets of global benchmarks may have been compromised because many companies/models might use them while competing to do well on these benchmarks - goodhart's law - still doesn't cover every dimension possible to judge a language model's capability. most of the models can exploit the statistical distribution to get answers on multiple-choice questions that most of these benchmarks have. we need more unit tests or subjective dimensions to benchmark new language models.

want to work on the intersection of ai, mathematics, consciousness, dna, memory, hardware

@n4gpal respect g

21 years old. just bought a frooti without emi. it ain't parent's money. it ain't luck. it ain't consistency. it's discipline. firm grind everyday welcomes this lifestyle.

@ShubhamTotu @cengines_studio you're getting better and better at this

“can it play doom?”

@thestoicccoder hell nah 😭

claude just killed Haldirams

The more I read about classical architectures (CNN, RNN, LSTM), the more I understand how we came to transformers and how, going from explicit hard-gated regex stuff to more hallucinative black-box stuff, we go towards a general intelligence. Kind of like we use our brain (hard-gated logic over experiences and rules) plus our real-time intuition(instinct, breaking the rules - hallucinating) in the real-world.

The evaluation pipeline for Indian ASR is broken. At Sarvam, we kept seeing WER/CER penalize correct output. They fail on: • Code-switching • Script variation (kal vs कल) • Same meaning, different phrasing So you get misleading scores → wrong product decisions. We built better metrics: LLM-WER, LLM-CER, Intent, Entity Evaluate Indian ASR in right way using these two frameworks.

you never know

@arcinston x.com/i/status/20338…

life lately

insane