Publications · For developers and engineering students
AI for Engineering
Building systems, not demos
A practical book on building AI systems that survive real use: retrieval that returns the right passage, agents with permission boundaries, evaluation that catches regressions, and deployment that does not fall over at the first traffic spike. Written from delivery work rather than from papers.
ICT Bangladesh
AI for Engineering
Building systems, not demos
- Writing
- Editing
- Printing
What's inside
What the book covers
Retrieval-augmented generation, and why most implementations fail
Designing agents with permission gates and audit trails
Evaluation sets — measuring quality instead of guessing at it
Working with Bangla: tokenisation, transliteration, evaluation
Self-hosting open models, and when it is the wrong choice
Cost, latency and capacity planning in production
Want a copy when it is out?
Send us a message and we will tell you when AI for Engineering is printed. Schools and institutions ordering in bulk are worth telling us about now, so we can plan the first run.
Contact Us