Home/Company
Company · Engineering Principles

A small engineering firm,
deployed inside complex operations.

Verne was founded on a simple bet: most organizations don't fail because they lack software, they fail because their operating logic is trapped across people, documents, and systems. We exist to make that logic visible, intelligent, and operationally reliable.

Founded
2024 · Dhaka
Engineers
32
Active Programs
12
Public Sector
6 countries
Why Verne Exists

Generic IT vendors fail at
this layer of the problem.

The work that matters in complex organizations sits in the seams. It lives between the ERP and the spreadsheet. Between the policy document and the caseworker. Between the field report and the quarterly briefing. Generic vendors treat each seam as a project. We treat it as a system.

Verne was built to be the firm that ships AI-native operating layers across those seams, production systems that are observable, auditable, and operated by your team after we step back. We are not a staff-augmentation shop. We are not a slide consultancy. We are forward-deployed engineers who happen to be very good at making AI work inside real organizations.

The vast majority of our work lives behind NDAs and inside regulated environments. The case studies you can see are a small public slice. What we ship behind the wall is what defines us.

Engineering Principles

Six principles that govern
how we ship.

01

Forward-Deployed by Default

Our engineers work inside your operation. We don't parachute decks, we ship code against your real systems and your real constraints.

02

Production over Prototype

We optimize for systems that run reliably in production, not demos that look good in pitches. Eval, observability, and audit are not afterthoughts.

03

Composable, Not Monolithic

Every system we build is composed of independently observable layers. We never deliver black boxes, your team owns and operates what we ship.

04

Human-in-the-Loop, Always

High-stakes decisions remain with humans. Our agents make humans faster, not replace their judgment. Approval and audit are first-class.

05

Boring Tech Where Possible

We default to the proven. Novel infrastructure is justified by the workload, not the resume. Production reliability beats architectural fashion.

06

Embedded, Then Independent

We engineer for skill transfer. Your team can operate, evolve, and extend the system after we step back. Forward-deployed, then forward-handed.

Leadership

Four founders.
Operators, engineers, policy-makers.

Verne was founded by four people who have spent their careers inside the systems we build for. Policy authorship at the national level. Operations at multilateral institutions. Engineering at hyperscale. Financial discipline across regulated industries.

Arefin Mizan, CEO, Verne
CEO

Arefin Mizan

Arefin works at the intersection of government, data, and AI. As a government technology consultant at the ICT Division of Bangladesh, he drafted the national AI policy, the national blockchain policy, and the national cloud policy, and served on the drafting committee for the national startup policy. Before that he was Senior Research Officer and Data Scientist at icddr,b, where he built large-scale simulation models for health, agriculture, and food systems.

  • Drafted national AI, blockchain, and cloud policies
  • Startup policy drafting committee
  • Senior Research Officer / Data Scientist, icddr,b
  • Simulation models across health, agriculture, food systems
Sharmin Jahan Juha, COO, Verne
COO

Sharmin Jahan Juha

Sharmin runs Verne operations. She brings deep enterprise-operations experience from the International Labour Organization and CARE, with a track record across digital transformation, ERP rollouts, procurement, partnerships, and fund management. Her domain is the unglamorous work that makes complex programs actually deliver.

  • Operations at the International Labour Organization (ILO)
  • Operations at CARE
  • Digital transformation and ERP rollouts
  • Procurement, partnerships, fund management
Sheikh Faisal Rahman Avash, CTO, Verne
CTO

Sheikh Faisal Rahman Avash

Avash leads Verne engineering. Previously a software engineer at AWS S3 and at Samsung R&D, and currently working at Tezos. He has designed and shipped large-scale AI and distributed systems at hyperscale, and brings that bar to every system Verne builds.

  • Ex-AWS S3 Engineer
  • Samsung R&D
  • Engineer at Tezos
  • Large-scale AI and distributed systems
Mansurul Kafi, CFO, Verne
CFO

Mansurul Kafi

Mansurul runs Verne finance. A Chartered Accountant in training (part-qualified), he has worked across audit, finance, and tax engagements in diverse sectors, and brings disciplined financial controls to a young company that will soon operate across multiple jurisdictions.

  • Chartered Accountant (part-qualified)
  • Cross-sector audit, finance, tax
  • Multi-jurisdiction financial controls
Security & Quality

Engineered for the most
regulated environments we serve.

The security posture you can expect by default. Engagements adjust this based on jurisdiction, sector, and the specific workload, but these are the baselines, not aspirations.

Identity & Access

  • SSO / OIDC / SAML
  • Role-based access control
  • Just-in-time elevation
  • Hardware key enforcement

Data Posture

  • Encryption at rest & in transit
  • Tenant isolation
  • Data residency
  • Key escrow for sovereign deployments

AI Risk Controls

  • Grounding & retrieval boundaries
  • Output validators
  • Prompt-injection defense
  • Continuous eval & regression

Audit & Compliance

  • Append-only audit logs
  • SOC 2 posture
  • ISO 27001 / 27701 alignment
  • NIST 800-53 control mapping
Careers

We hire engineers who want to ship
real systems against real constraints.

Not framework chasers. Not prompt theorists. Engineers who can sit across the table from a minister, a director of operations, or a compliance officer, and build the system they actually need.

AI Systems Readiness Audit

Bring us your
most complex workflow.

In 7–10 working days, Verne maps your workflows, data sources, repetitive decisions, automation opportunities, and AI risk areas. You receive a prioritized roadmap showing what to automate, integrate, avoid, and build first.

Tell us what is broken. We will map the system.