Alibaba Open-Sources AI Code Review Tool That Reduces Token Costs by 90%

·12 views

Alibaba has open-sourced Open Code Review (OCR), an AI-powered code review tool it has used internally for two years across tens of thousands of developers. The tool addresses three common pain points with existing AI review integrations: incomplete file coverage, inaccurate line-number positioning, and high token costs. OCR uses a hybrid architecture that separates deterministic tasks — such as file selection, bundling, and line tracking — from semantic reasoning, which is handled by large language models. Related files are grouped into bundles and reviewed by isolated sub-agents concurrently, reducing token usage by up to 90% compared to general-purpose AI agents. The tool is designed to integrate into both terminal workflows and CI pipelines.

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)

Developer Builds Independent APK Platform to Bypass Play Store Restrictions

A developer has launched Aeroapk (aeroapk.store), an independent Android app catalog designed as an alternative to the Google Play Store. The platform allows developers to publish their APKs directly to users, with each listing sourced exclusively from the original app creator. All submissions undergo a safety review before approval, and approved listings remain live until the developer decides to update or remove them. The project was motivated by common frustrations with traditional app stores, including slow review cycles, opaque ranking algorithms, and sudden policy changes. The creator plans to expand the catalog and refine the review process based on developer feedback as the platform grows.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Developer rewrites 2150-line Bash agent in Python, finds line count actually rose

A developer rebuilt a routine-task automation agent from 2150 lines of Bash across seven files into Python, ending up with 2258 lines — not the reduction most would expect. The decision to switch was not driven by performance, since the agent's runtime is dominated by AI model calls, not the 162 shell forks. The real trigger was an expressiveness problem: complex data manipulations required awkward jq subprocess calls that Python list comprehensions handled far more cleanly. A key architectural choice — storing prompts as standalone Markdown files with placeholders — meant no prompt content needed to change during the migration, keeping the rewrite contained to a single day. The developer concluded Bash was not the wrong original choice, but that a shifted requirement around data handling ultimately made Python the better fit going forward.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Understanding Python's range Function Is Key to Mastering Its for Loops

A developer with a C programming background shares insights on how Python's for loops differ fundamentally from those in C. Unlike C, Python's for loop is designed to iterate over collections, and numeric iteration is achieved through the built-in range function. The range object accepts up to three arguments — start, stop, and step — and occupies O(1) space regardless of the range size. Converting a range to a list helps visualize the sequence it represents, including odd numbers, even numbers, or reverse sequences. The author notes that while Python's for loops are more versatile than C's, understanding range is essential to using them effectively.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Anthropic Launches Invite-Only Life Sciences Program for Restricted Mythos 5.1 Access

Anthropic has introduced the Life Sciences Verification Program (LSVP), an invite-only beta that gives vetted life sciences researchers controlled access to its restricted Mythos 5.1 AI model. The program was developed in partnership with the US government and reduces biology-related restrictions for eligible participants while keeping safeguards intact in other areas such as cybersecurity. Mythos 5.1 is accessible through two trusted-access pathways: LSVP for life sciences researchers and the Cyber Verification Program for cyberdefenders. Initial participants have already been enrolled, with Anthropic planning to expand access to more qualifying life sciences organizations over time. Pricing for Mythos 5.1 under LSVP has not been publicly confirmed, and a 30-day data-retention policy applies to access under the program.

0 comments Read more at DEV Community