AI/ML Engineer + Fullstack Developer

Fahad Hossain

I pretrain Bengali language models from scratch and build the web applications that ship them.

01 · About

Self-taught.

I'm a developer from Dhaka. I started writing code in 2018, self-taught, on a phone. Web first, deep on SvelteKit and Go, then in mid-2026 I moved into LLM pretraining and trained Alo-70M from scratch on TPU v5e-8.

I've been obsessed with AI since GPT-2 was the biggest open-source model around. The goal is simple: bilingual Bengali models that are open weights, self-funded, and small enough to run anywhere. The web work pays for it and ships it.

Currently pretraining a bilingual LLM in the 1B parameter range . Self-funded, fully open source.

Facts

Location
Dhaka, Bangladesh
Focus
Bengali NLP · web
Coding since
2018
Founded
SurjoLabs (non-profit)
Training
Self-taught

Stack

Web

  • · SvelteKit
  • · TypeScript
  • · Tailwind CSS
  • · Go
  • · PocketBase
  • · Vite

ML

  • · PyTorch
  • · transformers
  • · Hugging Face Hub

Infrastructure

  • · Docker
  • · Netlify
  • · Vercel
  • · Git
  • · Linux

02 · Work

Selected work

Flagship

Alo-70M

69M Bengali LLM · Apache 2.0

69M-parameter Bengali LLM pretrained from scratch on TPU v5e-8. SFT on a curated 37.5K instruction mix. Outperforms Gemma-3-270M on two Bengali benchmarks. Apache 2.0.

PretrainingSFTBengali NLPEdge AI

02

SurjoLabs

Non-profit organization for bilingual Bengali models. Open weights, no funding, no paywalls.

OrganizationOpen source

03

PaymentTracker

In-production education management platform for a coaching center. Payments, students, and daily operations, beyond just tracking fees.

ProductionSaaSTypeScript

04

Pollin Coder

Free AI coding website. Maintainer: led the BYOP auth migration, feature flags, and moderation tooling. Powered by Pollinations.ai with Sandpack live sandboxes.

AINext.jsOpen source

05

Pathway

Turns an Android phone into a local SMS and USSD gateway. No cloud, no server, no subscriptions. Pure device-side networking.

AndroidNetworkingSvelte

06

Govt Photo Resizer

Client-side tool for Bangladesh government job and admission photo standards. 300×300 photos at 100KB, 300×80 signatures at 60KB, with iterative auto-compression. Nothing leaves the browser.

SvelteUtilityPrivacy

07

Omega-0

Design-focused web application built with attention to interaction and full-stack detail.

Web appFull stack

03 · ML model card

Alo-70M, evaluated

Zero-shot, continuation-based log-probability evaluation across seven Bengali reasoning and knowledge benchmarks. Every number below is published in the model card. The 69M-parameter model beats the 270M-parameter Gemma baseline on two of seven.

BenchmarkAlo-70M(SFT)Gemma-3-270MTigerLLM-1BAlo-70M-Base
bangla_mmlu_bn26.29%26.81%27.66%26.31%
bangla_commonsenseqa_bn25.88% ▲22.77%25.14%28.42%
indicbench_arc_bn_challenge24.15%25.34%27.13%22.70%
boolqa_bn48.70%51.30%52.40%48.42%
openbookqa_bn30.58%31.99%34.21%31.39%
piqa_bn50.05% ▲49.51%49.51%50.49%
hellaswag_bn26.89%27.85%31.01%27.27%

Alo-70M vs Gemma-3-270M

bangla_mmlu_bn

26.3
26.8

bangla_commonsenseqa_bn

25.9
22.8

indicbench_arc_bn_challenge

24.1
25.3

boolqa_bn

48.7
51.3

openbookqa_bn

30.6
32.0

piqa_bn

50.0
49.5

hellaswag_bn

26.9
27.9

Wins on bangla_commonsenseqa_bn and piqa_bn.

Size gap

3.9x fewer parameters than Gemma-3-270M

Wins

2 of 7 benchmarks, at 69M

Inference

CPU and phone, no GPU needed

Architecture

Layers
12
Hidden
512
FFN
1408
Attention
GQA 8 / 4
Positional
RoPE
Embeddings
Untied
Context
1024
Parameters
69M

Training

Hardware
TPU v5e-8
Optimizer
Fused AdamW
LR schedule
Cosine, peak 3e-4
Epochs
3
Batch
32 (grad-acc 4)
Precision
AMP
Format
ChatML
License
Apache 2.0

Tokenizer

  • Custom Bengali BPE, built for compact subword coverage.
  • ChatML formatting via the tokenizer's chat template.

Alignment tax

SFT aligned the model for stable generation and instruction following, but it cost measurable zero-shot reasoning: CommonsenseQA dropped from 28.42% to 25.88%. Structured-formatting samples (12.5K) were deliberately excluded from the mix; they crushed a sub-100M model's capacity. At 69M parameters the model is built for text processing (editing, summarizing, extraction), not encyclopedic retrieval. It runs on CPUs and phones.

Currently pretraining a bilingual LLM in the 1B parameter range

Self-funded, fully open source

04 · Journey

Hobbyist to pretraining

2018

First lines of code

Started writing code on a phone, after school. Self-taught, building whatever came to mind.

2019 · 2020

The AI bug

Got hooked on LLMs when GPT-2 was the biggest open-source model around.

Sep 2024

First portfolio ships

Self-taught web development. Started with SvelteKit and never looked back.

2025 · Mid

Fullstack deep-dive

Go, PocketBase, TypeScript. Shipped experimental apps and utility tools for the local market.

Nov-Dec 2025

First utility products

attendly, OCR, starter templates. Small tools with real users.

Jan-Feb 2026

Pathway + Pollin Coder

Built a serverless Android SMS gateway and became maintainer of a free AI coding platform.

Jun 2026

Govt Photo Resizer

A client-side tool for Teletalk photo standards. Thousands of applications prepared locally in the browser.

Jun-Jul 2026

The LLM era

Pretrained Alo-70M-Base from scratch on TPU v5e-8, SFT-aligned it on a curated Bangla instruction mix. It beats Gemma-3-270M on 2 of 7 Bengali benchmarks.

Aug 2026

SurjoLabs + 1B scale

Founded SurjoLabs, a non-profit for bilingual Bengali models. Started pretraining a bilingual 1B-parameter LLM, self-funded and open. PaymentTracker went live in a coaching center.

Fahad delivered exceptional work on our projects. His attention to detail and technical expertise with SvelteKit made all the difference.

Mats Dale Soma

Founder · Redveil

05 · Contact

Let's build something

LLM collaboration, SvelteKit work, Bengali NLP research, or just to talk shop. All channels below are open.