Work With Erik
Senior technical judgment for companies building difficult systems.
Direct help with systems that carry real consequences.
Erik Stenman, also known as Happi in the Erlang and BEAM community, helps teams reason about architecture, platforms, runtime behavior, technical leadership, and AI-assisted development.
HappiHacking is the independent technical imprint and systems work of Erik "Happi" Stenman: books, essays, courses, experiments, tools, and occasional advisory work. The advisory part is direct and personal. The work fits best when the system is difficult enough that experience matters.
When Erik Can Help
The useful fit is usually a company with real technical pressure: a platform that has outgrown its old shape, a payment or regulated workflow that needs better guarantees, a leadership gap, a BEAM system that needs runtime-level judgment, or an AI-assisted engineering process that needs traceability and control.
The System Is Hard To Explain
Architecture, ownership, queues, data flow, failure modes, and runtime behavior have drifted apart. The next decision needs a better model of the actual system.
Leadership Needs Technical Depth
Founders, executives, architects, or engineering managers need senior judgment during scaling, restructuring, platform change, or a gap between technical leaders.
Delivery Needs Control
AI-assisted engineering, regulated workflows, payment flows, or platform work need traceability, clear ownership, and operational discipline from the start.
Common Modes of Work
These are practical ways the work tends to happen. Scope depends on the system, access, urgency, and the people who need to own the result.
Architecture Review and Difficult-System Debugging
A focused look at code paths, runtime behavior, queues, state ownership, incidents, data contracts, and unclear decisions.
Fractional CTO, Interim CTO, Technical Director
Senior technical leadership during scaling, transition, restructuring, platform modernization, or a gap between technical leaders.
Technical and Architectural Coaching
Decision support for architects, tech leads, EMs, founders, and senior engineers who need clearer ownership and reasoning.
AI-Assisted Engineering
AI-assisted delivery with reviewable specs, traceability, governance, and operational ownership.
Platform Engineering and Reliability
Platform strategy, developer experience, observability, service boundaries, operations, and stable delivery.
BEAM Runtime Workshops
Private Erlang and Elixir workshops on processes, schedulers, memory, tracing, supervision, and production behavior.
Fintech and Regulated Systems Advisory
Technical judgment for money movement, risk, auditability, ledgers, payments, custody, and regulated workflows.
Recent Relevant Work
This is selected assignment history, grouped by the kind of buyer problem it helps explain.
Regulated AI-Assisted Delivery
SEB Financing / Brightshift
Spec-Driven Development and agentic AI-assisted delivery in regulated finance, with attention to governance-aware workflows, compliance integration, and traceable development processes.
Platform Engineering Leadership
Kivra
Interim Director of Platform Engineering and advisor to the CTO, working across architecture, reliability, developer experience, operations, strategy, and a BEAM-heavy platform.
Architect Coaching and Payment Flows
Redcare Pharmacy
Coaching architects and senior engineers on platform design, reliability, service boundaries, observability, and payment-flow modernization in a regulated pharmacy and e-commerce setting.
Vehicle-to-Cloud Platform Architecture
Scania / TRATON
Architecture work for a unified vehicle-to-cloud connectivity platform covering data exchange, integrations, diagnostics, OTA updates, and smart-fleet foundations.
Interim Technical Leadership
Laya, Kindio, Ignite Next
Interim CTO work, founder and CEO technical leadership, and mentoring for founders making early architecture, product, and team-shape decisions.
AI Compiler and Runtime Infrastructure
SambaNova Systems
Compiler and runtime infrastructure work around ML workloads, model integration, C++, LLVM, MLIR, PyTorch, TensorFlow, and build systems.
Proof Points
A few anchors explain the background. The full archive lives in systems work, books, talks, and endorsements.
Klarna CTO and Chief Scientist
First CTO at Klarna, later Chief Scientist. Built the engineering organization from one person to about eighty and worked on the technical foundation for a 24/7 payment system.
BEAM Runtime Work
The BEAM Book, PhD work at Uppsala University, HiPE, and long-running work around Erlang runtime implementation and production behavior.
Deutsche Telekom 1B+ Events Per Day
Architecture and proof-of-concept work for GDPR-aware event and IoT platforms capable of more than one billion daily events.
Aeternity FATE Runtime
Core team and architecture work on BEAM-based blockchain infrastructure, including FATE, a verifiable virtual machine for smart contracts.
Fit
Good Fit
- A system, platform, payment flow, or AI-assisted process has real consequences.
- The team can expose code, architecture, incidents, logs, metrics, or decision context.
- The goal is clearer technical ownership and better decisions by the internal team.
- Leaders want senior technical judgment with direct access to the person doing the work.
Poor Fit
- A large delivery team is required.
- The main need is a formal compliance certificate.
- The problem is generic process coaching.
- The real system, team context, or constraints cannot be shown.
Contact
Start with a short note about the system, the constraint, the decision you are facing, and what kind of access is possible.
