Applied ML · Dhaka, Bangladesh

The right
AI systems
for the right
170 million.

Six production AI systems — satellite flood intelligence, contract risk auditing, admissions equity, agricultural advisory, cricket analytics, industrial inspection. All built on free-tier infrastructure. All designed around the actual constraints of South Asia, not Silicon Valley.

6+
Production systems
$0
Cloud spend to date
170M
People in scope
78%
Of users offline
Satellite intelligence Contract risk auditing Admissions equity Agricultural advisory Cricket analytics Industrial inspection Disease forecasting Built on Google Cloud Satellite intelligence Contract risk auditing Admissions equity Agricultural advisory Cricket analytics Industrial inspection Disease forecasting Built on Google Cloud
§ 01 — The Thesis

Every country has
the same AI problems.
Only ours has
$1.60 solutions.

The standard playbook for "AI for development" is to take a mature Western system, translate it, and deploy it at scale. It fails because the infrastructure assumptions don't transfer. Bangladeshi garment factories don't have $12,000 optical inspection rigs. 78% of the rural population owns a button phone, not a smartphone. Government documents exist as degraded photocopies of mixed Bengali-English text.

My approach is different. I study the best-in-class solution in each domain, identify its infrastructure assumptions, and re-architect from first principles for the actual constraints of South Asia. That process consistently produces systems that are 10–80× cheaper to deploy while maintaining competitive accuracy — because the constraints force genuine innovation, not translation.

AI is the most powerful tool in human history that remains almost entirely unavailable to the people who need it most. That is a design problem. I build the designs.

"The insight from the garment factory: India sells ready-made optical inspection for crores. I rebuilt it on a recycled Android phone at $150 CAPEX. The factory doesn't care who made the component — it cares about the margin."

§ 02 — The Engineer

Self-taught.
Princeton-networked.
Production-focused.

I'm Samiul Karim — a self-taught applied ML engineer based in Dhaka, building production AI systems for domains that established labs ignore because the TAM looks small from San Francisco. From Dhaka it looks like 170 million people.

My background is unconventional: deep domain knowledge in garments manufacturing, Bangladeshi agriculture, public health surveillance, cricket analytics, and government document processing — combined with a Princeton University network built during an extended period as a Princeton alumni interviewer.

I work in long focused sessions. Every system in this portfolio was built to production-grade standards — proper authentication, eval harnesses, retry logic, secrets management — because demo-grade code teaches the wrong habits.

Reach me at sam@stitchmark.space

  • Princeton Alumni Network
    Former Princeton interviewer · active alumni connection
  • Toptal Application In Progress
    Applied ML Engineer track · portfolio submission
  • SPARRSO Partnership Discussions
    Bangladesh Space Research · FloodGuard deployment
  • Active Accelerator Applications
    M0NARQ Industrial AI · KrishiBot rural advisory
  • Six Production Systems
    Built at $0 cloud spend — not by accident, by design
  • GCP-Native Architecture
    Vertex AI RAG · Gemini API · Earth Engine · Cloud Run · Discovery Engine
§ 03 — The Work

Five systems.
Three verticals.
One city.

Each project below was built to production standards. None required paid cloud credits to reach a working prototype.

01
Legal Tech · RAG · Adversarial AI
DocRAG-Legal
Deployed ·
02
Admissions Intelligence · Bangladesh
Erudite V2
Pending Resources ·
03
Sports Analytics · BPL Cricket
CricSight Powerplay
Live ·
04
Geospatial AI · Disaster Response
FloodGuard-BD
In Development ·
05–07
AgriTech · Industrial AI · Public Health
The Lab — KrishiBot, M0NARQ, HAWKEYE Disease
Prototype ·
§ 04 — Google Cloud for Startups

Architecture validated.
Compute is the
only remaining gate.

All projects are technically proven on free-tier infrastructure. The engineering problems are solved. What scales impact is serving capacity — RAG corpus retrieval at volume, Gemini API at production query loads, Cloud Run autoscaling.

Projects are ranked below by social impact. Any credit allocation advances the work. Larger allocations unlock proportionally greater reach.

Impact Rank · 01 · Rural Agriculture
KrishiBot

AI agricultural advisory for 57 million rural Bangladeshis with no smartphone access — delivered over SMS and USSD at $0.001 per query. The constraint is not the AI. It is the last mile.

Minimum: $500 · Vertex AI RAG + Gemini Flash + Cloud Run
  • RAG corpus serving for BARC/DAE crop bulletins in Bengali
  • Gemini Flash API volume for SMS response generation at ~100k queries/month
  • Cloud Run serving for USSD/SMS gateway API at pilot scale

Greater allocation adds: full Bengali NLP pipeline, multi-crop seasonal corpus, carrier USSD deployment.

Impact Rank · 02 · Disaster Response
FloodGuard-BD

Near-real-time pixel-level flood maps from Sentinel-1 SAR for government disaster response. All-weather. No human annotation. No manual steps between satellite overpass and map delivery.

Minimum: $800 · Vertex AI Training + Prediction Endpoint
  • Vertex AI Training job for Prithvi-100M fine-tuning (single T4, ~2 hrs)
  • Vertex AI Prediction endpoint for real-time inference on new satellite passes
  • Cloud Storage for COG tile archive and public flood map output

Greater allocation adds: 4-month operational pilot, active learning loop with SPARRSO validation, public dashboard.

Impact Rank · 03 · Admissions Equity
Erudite V2

Forensic financial aid intelligence for high-need Bangladeshi students navigating elite US admissions — replacing generic chatbot advice with verified CDS audit data and structured visual counsel.

Minimum: $600 · Discovery Engine + Gemini Flash + Cloud Run
  • Discovery Engine corpus serving for 70+ page per-university deep research reports
  • Gemini Flash synthesis at production applicant query volume
  • Cloud Run serving for the applicant-facing Next.js frontend

Greater allocation adds: full corpus (15+ universities), Bangladesh-specific curriculum contextualization overlays.

Impact Rank · 04 · Legal Access
DocRAG-Legal

Contract risk intelligence with adversarial self-auditing. Bringing the evidentiary standards of a senior partner to every junior associate in Dhaka. Already deployed. Needs production serving scale.

Minimum: $700 · Vertex AI RAG Engine + Gemini API at volume
  • Vertex AI RAG Engine corpus serving at ~10k queries/month production load
  • Gemini Pro synthesis + Flash adversarial judge at concurrent volume
  • Cloud Run autoscaling (0–5 instances) for SaaS-tier user load

Greater allocation adds: Bangladesh-specific corpus, Bengali-language query handling, SaaS tier launch at $149/month.

Vertex AI RAG Engine Gemini Flash · Pro Google Earth Engine Discovery Engine Cloud Run Firebase Auth Modal Persistence SHA-256 Provenance Vertex AI RAG Engine Gemini Flash · Pro Google Earth Engine Discovery Engine Cloud Run Firebase Auth Modal Persistence SHA-256 Provenance