Harten Papers
Articles examining how teams understand existing applications, agree which behaviour to preserve and check AI-assisted software changes.
Each paper develops a specific question, with examples and source references. Browse by title below; the date identifies when the piece was published.
The Modernisation Control Papers · 06.2
Compare possible starting points for a new application by the uncertainty they resolve, the useful software they leave behind and the next engineering decision their results can support.
Harten Technical Papers · Technical report: sustained agentic engineering
Wayfinder was built to analyse complex applications with low token consumption. Creating and refining that capability involved approximately 95 billion tokens of Codex and Foundry usage. Source-analysis projections based on 316 recorded results put the eight current applications, covering 15.2 million source lines, at 500 million tokens: around 190 times smaller than the engineering investment. A rerun of all 40.76 million recorded lines at the current workflow's efficiency projects to 1.34 billion tokens. This paper explains the upfront work behind that repeatable analysis capability, connects the experience to published guidance and develops twelve practices other teams can apply across planning, review, recovery, coordination and delivery.
The Modernisation Control Papers · 06.1
Follow a payment decision through its data, operational holds and downstream effects to judge the real scope of a change, rather than using the size of its implementation as a measure of risk.
The Modernisation Control Papers · 06
A completely new application can be built through a sequence of bounded engineering work. This paper examines how consequences, dependencies, useful learning and recovery should shape that sequence.
The Modernisation Control Papers · 05.3
Build a completely new application from agreed business requirements, then keep its specification, application knowledge and verification current so engineers can continue improving or replacing the implementation.
Harten Technical Papers · Technical report: engineering agents
AI changes how engineers work and how they learn. Junior developers can contribute useful engineering now, acquire judgement through that work and use AI to participate more fully. Removing the entry route would weaken the capability organisations still need.
The Modernisation Control Papers · 05.2
Two designs for an eligibility service show how to compare independent releases, consistent decisions, recovery and operating costs, and explain what would justify choosing a different architecture.
The Modernisation Control Papers · 05.1
Why an eligibility service might have three repositories, what a policy change reveals about their dependencies, and how engineers can use that evidence to choose boundaries for a new application.
Harten Technical Papers · Technical case analysis
Six PRSD-IWS examples show how source-linked findings help engineers preserve behaviour, investigate uncertainty and define meaningful tests for AI-assisted software change.
Harten Technical Papers · Technical paper — 06
A proposed evidence-linked model for assessing whether a modernisation plan can meet its budget, deadline and acceptance criteria, grounded in Harten's public PRSD-IWS analysis and one-week analysis objective.
The Modernisation Control Papers · 05
An eligibility-service example shows how evidence can support a greenfield replacement rather than recreating the legacy architecture, and why production transfer still requires independent verification and human approval.
The Modernisation Control Papers · 04.3
Migration verification should check the new system against the behaviour the organisation decided to preserve, including exceptions and deliberate changes, rather than assuming the old implementation is always the correct oracle.
The Modernisation Control Papers · 04.2
An order can appear cancelled in one system while the warehouse still dispatches it. Understanding that failure helps us specify what a replacement must guarantee without recreating the old implementation.
The Modernisation Control Papers · 04.1
Before rebuilding an application, teams need to decide which outcomes, rules and controls still matter and which parts of the current system should be allowed to disappear.
Harten Technical Papers · Technical paper — 05
An empirical Harten case study of software agents coordinating other software agents while durable doctrine keeps consequential decisions, scope changes and acceptance of risk with the human principal.
The Modernisation Control Papers · 04
A migration should preserve the behaviour the organisation still needs, not every detail of the implementation it is replacing.
The Modernisation Control Papers · 03.3
When AI contributes to a software decision, the useful output is not just the answer. Teams need to see what was found, what supports it, what remains unresolved and where a person still needs to decide.
The Modernisation Control Papers · 03.1
A fluent answer can look certain while missing the scope, dependencies and contradictions that determine whether a software claim is actually useful.
The Modernisation Control Papers · 03
AI systems can produce convincing answers without establishing why those answers should be trusted. Evidence connects claims to sources, scope, uncertainty and the decisions that depend on them.
Harten Case Studies · Case study 01
A public benchmark using NightingaleHQ's MIT-licensed mixed legacy example corpus: what MaaS could establish in approximately an hour, what it left unresolved, and why the gaps matter.
The Harten Papers · Empirical follow-on — 16 September 2026
A live engineering case linking sustained work, accountable change and continuous renewal beyond individual model contexts.
The Modernisation Control Papers · 02.03
Discovery should distinguish what an application does, what the organisation needs and what it has authorised someone to change. That distinction connects application understanding to a defensible modernisation decision.
Harten Technical Papers · Technical paper — 04
An empirical case study of evidence-bearing distributed termination: how autonomous engineering agents can establish that shared work is complete rather than merely report that their local task has finished.
Harten Technical Papers · Technical paper — 03
An empirical case study of autonomous software-engineering agents coordinating shared mutable state through explicit invariants, state-dependent evidence and minimum necessary serialisation.
Harten Research · Research 01
An evidence-directed approach to reconstructing complex software systems, compressing application discovery and directing human attention to the uncertainties that still require judgement.
Harten · Working observation
Roughly six hours of essentially autonomous engineering execution, with human review and decisions on direction. A working observation from building Harten.
The Modernisation Control Papers · 02.02
Legacy systems become expensive to change when the organisation loses reliable knowledge of how they actually work, where behaviour lives and what a change may affect.
Harten · Standalone
Long-running AI systems need durable state outside the model context if decisions, evidence and engineering intent are expected to survive compaction and continue reliably over time.
The Modernisation Control Papers · 02
What appears to be poor engineering productivity is often the cost of changing a system the organisation no longer understands well enough.
The Modernisation Control Papers · 01.3
Safe autonomous change depends on evidence-backed understanding across code, operations, people and policy. That understanding must preserve provenance, uncertainty and conflicting perspectives.
The Modernisation Control Papers · 01.2
Human oversight cannot scale by requiring a person to inspect every machine action. Human authority must scale by concentrating judgement where evidence, uncertainty and consequence require it.
The Modernisation Control Papers · 01.1
AI makes software production dramatically faster, but velocity does not establish whether a change is understood, justified or safe. As generation gets cheaper, confidence becomes scarce.
The Modernisation Control Papers · 01
AI is reducing the cost of producing software change. The scarce capability is increasingly the evidence, authority and control required to decide which changes are justified.