Consulting
Available for consulting on AI-augmented engineering practice, agent infrastructure, and data-access architecture for agentic systems. Engagements range from targeted advisory to hands-on architecture and implementation.
Data-Estate X-Ray
I show you what your database actually looks like — recovered from the database itself, not from your docs.
Recovered Entity-Relationship Map — Excerpt
- 73-table estate mapped in seconds: 72 business entities, 979 fields, 114 relationships — deterministic, reproducible, no AI guesswork on the map.
- 45% of monthly production rows silently drop when joined to the field master — the kind of defect every dashboard and AI agent on top of this database inherits invisibly.
- 86% of declared foreign keys were never validated by the database engine (NOT VALID). The schema documents far more integrity than the database guarantees — most tools cannot tell the difference. Ours can.
- 798 human-authored column descriptionsrecovered from the database's own COMMENT metadata and carried into every artifact — the estate's own documentation, surfaced instead of guessed.
- We name what we miss. The audit caught gaps in our own tooling — twice, both root-caused and fixed the same day. The report does not hide gaps — it flags them.
- Zero orphaned rows in all probed declared relationships. We report health, not just problems.
Offers
Agent-Readiness Assessment
Scored report: can your data survive an AI agent? Live blind-agent probes, silently-wrong detection, remediation list
Data-Estate X-Ray
Recovered semantic model + narrated estate + defect register + ER map + provenance certificate
Semantic Drift Watch
Monthly re-introspection + meaning-level diff report + alert on change
Pricing pending final confirmation. Contact to discuss your estate.
See the full sample report and field-level data dictionary — a real Data-Estate X-Ray run on the Norwegian Petroleum Directorate FactPages (73 tables, public data).
Consulting Practice
AI-Augmented Engineering
Agent harnessing, human–agent development methodologies, standards enforcement via agents, long-horizon agent memory and retrieval
Data & Query Systems
Automated schema discovery, ontology inference, natural-language to structured-query translation, graph and vector stores, benchmark-driven evaluation
LLM Integration
Multi-provider abstraction for local and hosted models, prompt caching, tool use, structured outputs, agent protocols (MCP)
Languages & Frameworks
Python, SQL, Bash · Django, Flask, FastAPI
Machine Learning & CV
PyTorch, TensorFlow/Keras, scikit-learn · OpenCV, YOLO · NumPy/SciPy/Pandas stack
Cloud & Infrastructure
Docker, AWS, self-hosted CI/CD, containerized deployment pipelines
General inquiry
Tell me about the problem you are trying to solve. I will be honest about whether I can help and how.
Start a conversation