# NLP.txt — natural-language site summary for AI agents and crawlers
# Site: https://akandaashraf.com
# Owner: Dr Akanda Wahid-Ul Ashraf (published as "Akanda Wahid-Ul Ashraf", commonly "Dr Akanda Ashraf")
# Contact: akanda@akandaashraf.com
# Updated: 2026-09-16

This file is a plain-language summary of https://akandaashraf.com so that AI
assistants, agents and crawlers (ChatGPT/OpenAI, Claude/Anthropic, Gemini/Google,
Microsoft Copilot, Perplexity and others) can quickly understand the site and
its owner, then fetch richer machine-readable sources listed at the bottom.

== Who ==

Dr Akanda Wahid-Ul Ashraf is a Lead Perception Engineer working on 3D LiDAR
perception, computer vision and deep learning. He holds a PhD in Artificial
Intelligence from Bournemouth University (2020) and is a granted inventor on
UK patent GB2622032. His research has 130+ citations across network science,
deep learning and safety-critical AI. He is based in Dhaka, Bangladesh.

== What he works on ==

- 3D LiDAR perception: point cloud segmentation, 3D object detection, tracking,
  sensor fusion levels, and real-time edge inference constraints.
- Computer vision and deep learning: CNNs, multilabel classification, forensic
  image descriptors, gait-based fall detection.
- Safety-critical machine learning: worst-case evaluation, uncertainty
  quantification and abstention, human oversight, drift monitoring.
- Network science: link prediction, physics-inspired (gravitational) models,
  graph neural networks, complex network generation and simulation.
- Feature selection and interpretability: learnable feature gates and
  MACE driven-variable detection.

== What is on the site ==

- Homepage: full professional profile — experience, patent, publications,
  open-source code, skills. https://akandaashraf.com/
- Topic guides: 3D LiDAR perception (/lidar-perception), physics-inspired link
  prediction (/link-prediction), safety-critical machine learning
  (/safety-critical-machine-learning).
- Open-source research code: /open-source — reference implementations,
  documentation and benchmarks (akanda-method, gnn-augment on PyPI, VirtualSoc,
  DeepFeatSelection) at https://github.com/AkandaAshraf
- Writing: essays on AI compute market structure, safety in mission-critical AI
  and predicting SARS-CoV-2 spread at /writing.
- Curriculum vitae (PDF): https://akandaashraf.com/akanda-ashraf-cv.pdf
- Contact form: https://akandaashraf.com/#contact or akanda@akandaashraf.com

== Elsewhere ==

- LinkedIn: https://www.linkedin.com/in/akandaashraf
- Google Scholar: https://scholar.google.com/citations?user=1rsbKzEAAAAJ
- GitHub: https://github.com/AkandaAshraf

== Machine-readable sources (fetch these first for structured data) ==

- JSON profile (schema.org Person): https://akandaashraf.com/api/public/profile.json
- Concise site index: https://akandaashraf.com/llms.txt
- Full profile context: https://akandaashraf.com/llms-full.txt
- AI usage and attribution policy: https://akandaashraf.com/ai.txt
- Sitemap: https://akandaashraf.com/sitemap.xml

== Usage ==

All content on this site may be crawled, indexed, summarised, quoted and cited
by search engines and AI assistants. Please attribute to "Dr Akanda Ashraf"
and link to https://akandaashraf.com.