Raihan

404 bio not found

Bangladesh Joined on Oct 21, 2019 https://raihan-js.github.io

Education

Computer Science

Work

Founding Engineer @ VETR Proposal

Raihan

Oct 10

Teaching a Computer to Hear Japanese Pitch Accent

#llm #machinelearning #python #japanese

2 min read

Want to connect with Raihan?

Create an account to connect with Raihan. You can also sign in below to proceed if you already have an account.

Create Account

Already have an account? Sign in

Raihan

Oct 10

I Distilled a 568M Multilingual Model Into a 37M Japanese-English Encoder — Here's What Survived

#machinelearning #nlp #python #japanese

4 min read

Raihan

Oct 9

A Grammar Stopped My Model Inventing FAR Clauses. It Also Made It Refuse the Real Ones

#llm #machinelearning #python #nlp

4 min read

Raihan

Oct 8

Mentioned Is Not Binding: Which Solicitation Clauses Actually Bind the Contract

#llm #machinelearning #python #nlp

3 min read

Raihan

Oct 8

GraphProof-QA: Teaching a Small Model to Prove Its Answers

#llm #machinelearning #python #nlp

4 min read

Raihan

Oct 7

Small LLM Judges Approved 11% and 41% of Wrong Answers. Then I Fixed My Own Pairwise Test.

#llm #machinelearning #python #evaluation

5 min read

Raihan

Oct 6

Do LLMs Invent Japanese Law Articles? A Bilingual Benchmark

#llm #nlp #japanese #opensource

4 min read

Raihan

Oct 6

kube-memtier: Measuring Per-Pod Cold Memory with DAMON

#kubernetes #linux #monitoring #performance

4 min read

Raihan

Oct 5

AWQ Looked 10 Points Better on GSM8K Until I Stopped Truncating the Answers

#llm #machinelearning #python #testing

12 min read

Raihan

Oct 4

How Accurate Is Your Model Right Now? Estimating Accuracy Without Labels

#machinelearning #mlops #python #monitoring

4 min read

Raihan

May 12

I built the first open benchmark for federal contracting AI. Here's what it shows about frontier LLMs.

#ai #machinelearning #nlp #opensource

6 min read

Raihan

May 12

Three small models for healthcare intake — and what shipping all three taught me

#machinelearning #healthcare #ai #python

6 min read

Raihan

May 12

Where small models beat frontier LLMs (and where they don't): a 125M PHI detector

#machinelearning #healthcare #ai #python

10 min read

Raihan

May 11

Matching frontier LLMs at 22 lower latency: a 184M-parameter intent classifier for healthcare text

#ai #machinelearning #python #huggingface

9 min read

DEV Community

Raihan

Writing Debut

Six Year Club

Five Year Club

Four Year Club

Three Year Club

Two Year Club

One Year Club

Skills/Languages

Teaching a Computer to Hear Japanese Pitch Accent

Want to connect with Raihan?

I Distilled a 568M Multilingual Model Into a 37M Japanese-English Encoder — Here's What Survived

A Grammar Stopped My Model Inventing FAR Clauses. It Also Made It Refuse the Real Ones

Mentioned Is Not Binding: Which Solicitation Clauses Actually Bind the Contract

GraphProof-QA: Teaching a Small Model to Prove Its Answers

Small LLM Judges Approved 11% and 41% of Wrong Answers. Then I Fixed My Own Pairwise Test.

Do LLMs Invent Japanese Law Articles? A Bilingual Benchmark

kube-memtier: Measuring Per-Pod Cold Memory with DAMON

AWQ Looked 10 Points Better on GSM8K Until I Stopped Truncating the Answers

How Accurate Is Your Model Right Now? Estimating Accuracy Without Labels

I built the first open benchmark for federal contracting AI. Here's what it shows about frontier LLMs.

Three small models for healthcare intake — and what shipping all three taught me

Where small models beat frontier LLMs (and where they don't): a 125M PHI detector

Matching frontier LLMs at 22 lower latency: a 184M-parameter intent classifier for healthcare text