Developer Builds AI Exam Prep App in 8 Months After Frustration with Rote Study Methods

·1 views

A computer science student built ExamIntelligence, an AI-powered exam preparation app, after growing frustrated with exams that reward pattern recognition over genuine learning. The project began as a rushed, vibe-coded prototype using the Gemini API and Streamlit just days before his preliminary exams. After prelims, he discovered the AI-generated codebase was riddled with errors, prompting a full rebuild from scratch in Neovim with a more disciplined, architecture-first approach. He developed a hybrid AI pipeline, benchmarking multiple PDF parsers and testing local language models before settling on a multimodal solution to handle complex documents. The app, now live at examintelligence.app, aims to parse past papers and mark schemes to help students study more efficiently and free up time for deeper, curiosity-driven learning.

Read the full story at DEV Community

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

RFC 1071 Checksum in x64 Assembly: How ICMP Packets Are Validated

A developer working on an ICMP-based network communication project has shared a detailed breakdown of the RFC 1071 checksum algorithm implemented in x64 Assembly. The checksum is a 16-bit one's complement sum that ensures data integrity by processing a packet in 2-byte chunks and summing them. If the checksum is incorrect, the receiving operating system discards the packet as malformed before it is processed. The author walks through the Assembly code step by step, explaining how registers are used to iterate over the data buffer and handle odd-length packets. The piece frames the exercise as a practical way to understand low-level memory layout and CPU byte processing.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

MVP Development Typically Takes 4 to 12 Weeks, Scope Is the Key Factor

Most software startups can realistically build a minimum viable product in 4 to 12 weeks, depending on complexity, according to a development framework breakdown. Founders are advised to plan timelines around project phases rather than feature lists, as scope ambiguity is a leading cause of delays. A basic SaaS MVP covering core workflows may take 4 to 8 weeks, while more complex builds involving AI, marketplaces, or advanced integrations can exceed 12 weeks. Experts warn that skipping discovery, UX planning, or QA testing often slows development or damages early user experience. Ultimately, MVP success should be measured not just by build speed but by how quickly the product generates actionable learning from real users.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Celebrity Problem: Find the Party Celebrity in O(N) Using Two Pointers

The Celebrity Problem is a classic coding interview question where a celebrity is defined as someone who knows no one but is known by everyone, represented in an N×N matrix. A brute-force approach checks every row and column individually, yielding O(N²) time complexity. A more efficient solution uses a two-pointer elimination strategy: if person A knows person B, A is eliminated; if A does not know B, B is eliminated. After N-1 comparisons, one candidate remains and is verified by checking their full row and column. This optimized approach reduces time complexity to O(N) while maintaining O(1) space.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Developer Builds ICMP Packet Sniffer in x64 Assembly Without Any C Libraries

A developer has published an open-source ICMP network sniffer written entirely in x64 Assembly, bypassing standard C libraries and relying solely on direct Linux system calls. The tool uses a raw socket via syscall 41 with AF_INET and SOCK_RAW parameters to capture only ICMP packets, filtering out all TCP and UDP traffic at the OS level. Incoming packets are manually parsed by stripping the 20-byte IPv4 and 8-byte ICMP headers to access the raw payload. A custom integer-to-ASCII engine, built without any external functions, converts binary IP addresses into human-readable strings using division instructions and ASCII offset calculations. The project, released on GitHub under the handle JM00NJ, is intended strictly for educational purposes and security research on networks the user owns or has permission to monitor.

0 comments Read more at DEV Community