Developer Builds Ethical AI Voice Agent to Handle Subscription Cancellations

·1 views

A developer has released an open-source AI-powered subscription cancel-save agent built with Python Flask, Telnyx Voice, and AI Inference. The agent classifies a caller's cancellation reason and sentiment, then offers a single relevant retention option rather than pressuring customers with multiple prompts. Hardcoded overrides ensure that urgent phrases like 'fraud' or direct cancellation requests immediately bypass any save attempt, preventing manipulative dark patterns. Unresolved calls due to hangups are flagged for human follow-up, and all outcomes are logged for tracking. The demo code is publicly available on GitHub and is designed to integrate with billing platforms like Stripe or Chargebee in a production environment.

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 JA4H HTTP Fingerprinting for Amazon CloudFront to Block Bots

A developer has implemented a JA4H-equivalent HTTP fingerprint using Amazon CloudFront Functions, filling a gap left by CloudFront's native support for TLS-based JA4 but not HTTP-layer fingerprinting. JA4H is part of the JA4+ suite created by FoxIO, and generates a fingerprint from HTTP request headers to help identify and distinguish client implementations. While CloudFront added support for JA3 in 2022 and JA4 in October 2024, the HTTP-based JA4H variant requires a custom implementation via CloudFront Functions. The solution uses the CloudFront-Viewer-Header-Order header to compute a JA4H-equivalent fingerprint entirely at the edge, without modifying origin infrastructure. Combined with CloudFront KeyValueStore, operators can maintain and update a blocklist of suspicious fingerprints in real time without redeploying any code.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Federal Procurement Shift Favors AI-Native Firms Delivering Fixed-Price Outcomes

The U.S. federal government is overhauling its procurement model, moving away from labor-hour contracts toward fixed-price, performance-based agreements under initiatives like the Revolutionary FAR Overhaul. This structural change advantages AI-native Outcome Integrators — firms that combine domain expertise, automation-first engineering, and managed-service delivery. These vendors have gained early traction by presenting working prototypes during the RFI phase and submitting bids estimated at 60–70% of incumbent budgets, enabled by automation and reusable components. Zero Trust and cloud-native architectures further support this shift by making secure, managed-service delivery feasible without agencies operating every production layer themselves. Procurement leaders are advised to reframe solicitations around measurable outcomes, request runnable prototypes during market research, and build knowledge-transfer requirements into contracts to avoid institutional knowledge lock-in.

0 comments Read more at DEV Community

ProgrammingHacker News ·

Developer Releases Tool to Prevent Swastika Patterns in QR Codes

A developer has published an open-source Rust crate called 'qr-swastika-avoider' on crates.io. The tool is designed to detect and avoid the accidental generation of swastika-like visual patterns that can appear in QR codes due to their black-and-white pixel arrangements. Such unintended patterns can cause reputational or social harm when QR codes are used in public-facing contexts. The library gives developers a way to regenerate or adjust QR codes that would otherwise render offensive shapes. The project was shared on Hacker News, where it received modest early attention.

0 comments Read more at Hacker News

ProgrammingDEV Community ·

E-Commerce Chatbot Score Rises from 86 to 91 After Targeted System Prompt Fixes

A team at BotCritic audited an e-commerce customer support chatbot by testing it against four simulated customer personas covering order tracking, returns, refunds, and product queries, scoring it 86 out of 100 on the first run. The audit identified four specific flaws: the bot asked for order numbers it could not actually look up, omitted item-condition requirements from return explanations, gave an ambiguous refund trigger, and failed to proactively warn users who may have exceeded the 30-day return window. Developers addressed all four issues by updating only the bot's system prompt, with no other changes to the underlying model or configuration. Re-running the identical test suite produced a score of 91 out of 100, with the largest improvement seen in the Robustness category, which rose from 78 to 88. The exercise highlights how precisely targeted prompt changes can meaningfully improve chatbot performance without requiring deeper technical overhauls.

0 comments Read more at DEV Community