35 ARTICLES · 6 THEMES · CODE EXAMPLES
Building MailTrace
How an evidence application connects deterministic rules, AI interpretation and human review—and how to evaluate the next capability before adding it.
Start with article 1 →Architecture & product
All articles in this theme →Capability ownership, understandable interfaces and shared web and Android workflows.
Engineering note
Building MailTrace: Turning Email Correspondence into a Searchable Evidence Workspace
How MailTrace connects correspondence, evidence, issues and saved answers without losing their source history.
Engineering note
Where Deterministic Code Ends and AI Begins in MailTrace
A practical boundary between model interpretation and the backend rules that protect source evidence.
Engineering note
Making AI Processing Understandable in the MailTrace Interface
Turn backend states into clear actions, saved progress and honest explanations for people working with evidence.
Architecture proposal
Using Pretrained AI Versus Training Models with TensorFlow or PyTorch
Explain where deep-learning frameworks could fit without implying that every AI product needs its own neural-network training stack.
Architecture proposal
Do MailTrace’s Workflows Need LangGraph, CrewAI, AutoGPT or Bee?
Evaluate orchestration frameworks against durable jobs, explicit tool permissions and the existing application workflow.
Architecture proposal
Designing One Evidence Backend for Web and Android
Share evidence rules and saved jobs across clients while giving Android an interface suited to mobile use.
Architecture proposal
What a Year of MailTrace Iterations Taught Me About AI Product Architecture
Why consolidation needs maintained capability owners, explicit contracts and workflow tests rather than another collection of copied modules.
Evidence & preparation
All articles in this theme →Source preservation, email features, attachments and document processing.
Engineering note
Feature Engineering for Email Evidence: Threads, Participants, Dates and Requests
How useful email features combine semantic text with explicit context, and why missing context should remain visible.
Engineering note
Cleaning Email Text Without Changing the Original Evidence
Use derived text for retrieval while preserving original messages, source hashes and a route back to the evidence.
Experiment design
Can a Small Classifier Separate Signature Assets from Useful Evidence?
Combine attachment signals without excluding a useful file solely because it is small.
Engineering note
From PDF to Searchable Evidence: Text Extraction, OCR and Multimodal Processing
Distinguish a stored file from searchable text, and expose unsupported media processing honestly.
Embeddings & retrieval
All articles in this theme →Semantic search, hybrid retrieval, grounded answers and citation validation.
Engineering note
Using OpenAI Embeddings to Find Related Evidence in MailTrace
Where text embeddings fit in MailTrace, how source chunks are indexed, and what similarity does not prove.
Engineering note
Why Embedding Dimensions and Model Versions Matter in Production
How explicit embedding contracts prevent incompatible indexes and make migrations reviewable.
Engineering note
Hybrid Retrieval in MailTrace: Combining Semantic Search with Keyword Matching
Why evidence search needs both paraphrase matching and exact terms, with a small rank-fusion example.
Engineering note
Building Evidence-Grounded Q&A with RAG and the Responses API
How scoped retrieval, structured answer proposals and saved artifacts fit together in MailTrace Q&A.
Engineering note
A Citation Is More Than a Link: Validating AI Answers Against Source Evidence
Validate source scope, freshness and exact quotations before treating an answer as evidence-grounded.
Experiment design
Reranking Retrieved Evidence Before Asking the LLM
A proposed second-stage relevance check for MailTrace retrieval, with clear limits on cost and authority.
Clustering & issue review
All articles in this theme →Candidate groups, distinct requests, broad families and reversible decisions.
Engineering note
Constrained Hierarchical Clustering for Email Evidence
How complete-link candidates and contextual constraints can reduce misleading issue merges.
Engineering note
Why Twelve Categories Do Not Mean Twelve Clusters
Keep broad navigation families separate from the number of concrete issues in a case.
Engineering note
Preventing False Merges: When Similar Emails Describe Different Requests
Why request references, dates and missing context must constrain semantic grouping.
Engineering note
Designing AI Suggestions That Users Can Review and Reverse
How proposal, apply and restore stages keep AI organisation changes understandable and recoverable.
Engineering note
The 69-Item Test That Exposed a Workflow Dead End
What a synthetic consolidation fixture revealed about bounded AI output, unassigned evidence and active issue visibility.
Experiment design
K-Means, DBSCAN and Hierarchical Clustering: Which Fits Evidence Organisation?
Compare clustering assumptions against the actual cost of merging or separating case evidence.
Experiment design
Could Gaussian Mixture Models Represent Ambiguous Evidence Groups?
Explore soft cluster membership without mistaking model responsibilities for evidence certainty.
ML experiments & evaluation
All articles in this theme →Classifiers, leakage prevention, reviewer labels and honest quality measurement.
Engineering note
Testing an AI Application: What Mocked Tests Prove—and What They Leave Unanswered
Separate deterministic regression tests, provider integration checks and real evidence quality evaluation.
Experiment design
Can Logistic Regression or Naive Bayes Classify MailTrace Messages Cheaply?
A proposed baseline experiment for narrow email triage, with a scikit-learn pipeline example.
Experiment design
Random Forest, XGBoost, LightGBM and CatBoost for Email Triage
Where tree ensembles might help structured email triage, and how to compare them without inventing a winner.
Experiment design
Cross-Validation for Email Data: Preventing Thread and Case Leakage
How to evaluate email models without training on near-duplicates of the evidence used to test them.
Experiment design
Learning from Reviewer Corrections in MailTrace
Turn reviewed decisions into a governed evaluation dataset before using them to train another model.
Experiment design
PCA and t-SNE for Exploring MailTrace’s Embedding Space
Use embedding visualisations to investigate patterns without treating a two-dimensional picture as a clustering verdict.
Experiment design
Isolation Forest and One-Class SVM for Evidence-Import Anomalies
Explore anomaly detection for import quality while keeping unusual evidence distinct from invalid evidence.
Experiment design
When Would Fine-Tuning Improve MailTrace?
Decide whether a measured failure belongs to retrieval, prompting, validation or model training before fine-tuning.
Production & efficiency
All articles in this theme →Durable jobs, saved results, observability and task-specific model selection.
Engineering note
Running AI Work in the Background with FastAPI and Persistent Jobs
Why a saved job lifecycle matters more than a spinner when AI work is slow, retryable and paid.
Engineering note
Reducing Repeated AI Work with Versioned Indexes and Saved Results
How reuse, explicit invalidation and saved artifacts reduce latency without hiding stale evidence.
Experiment design
Choosing Smaller Models for Routine Tasks in an Evidence Application
A task-specific routing experiment that weighs quality, latency and cost without using model size as a safety rule.