Hacker News

Article URL: https://www.scientificamerican.com/article/the-arecibo-message-earths-first-interstellar-transmission-turns-50/

Comments URL: https://news.ycombinator.com/item?id=42196881

Points: 2

# Comments: 0

Article URL: https://uxmovement.substack.com/p/badged-field-labels-a-better-variant

Comments URL: https://news.ycombinator.com/item?id=42196524

Points: 1

# Comments: 0

Article URL: https://langrocks.com/

Comments URL: https://news.ycombinator.com/item?id=42196518

Points: 1

# Comments: 0

Hey Hacker News! We’re Badr, Ismail, and Skandere, co-founders of Sandra AI (https://www.sandra-ai.com/en). Sandra AI is the first multilingual voice AI receptionist built specifically for car dealerships. It handles both inbound and outbound calls, helping dealerships capture every customer opportunity while freeing up their staff to focus on in-person interactions and closing deals.

Here’s a demo video: https://youtu.be/89BbgavAffQ

Phones are still a critical channel for car dealerships, with around 80% of service appointments booked by phone. Yet, 30% of calls typically go unanswered, leading to lost revenue and frustrated customers. Traditional call centers are expensive, hard to scale, and often fail to meet demand. Delays and missed follow-ups frequently drive customers toward competitors.

We designed Sandra AI to tackle these issues. Using multilingual voice AI, Sandra communicates naturally in multiple languages, integrates directly with dealership management systems (DMS), and operates 24/7. It ensures no call is missed and transfers calls to human staff when necessary, following dealership-specific rules. With full visibility into every interaction—recordings, transcripts, and satisfaction metrics—dealers can trust Sandra to improve both customer experience and operational efficiency.

The idea for Sandra AI came from our own experiences in the automotive and BPO industries. We saw firsthand the impact of missed calls on customer satisfaction and revenue. When we noticed how advancements in voice AI could address these challenges, we set out to build a solution tailored specifically to the needs of car dealerships.

Under the hood, Sandra AI is built on top of open-source libraries (like pipecat or vocode) and we use WebSockets with Twilio to manage phone calls. The biggest technical challenge we faced was integrating with the often legacy software of dealerships, which typically lack APIs. This required a significant amount of engineering work to ensure compatibility and smooth data flow between systems. In terms of pricing, we use a subscription model per dealership, with variations based on size and the volume of calls to be handled.

Building AI that feels truly conversational across languages and integrating seamlessly into dealership workflows hasn’t been easy, but we’re excited by the progress so far.

We’d love to hear your feedback! If you run a car dealership or know someone who does, we’d be thrilled to explore how Sandra AI can help. We’re also keen to discuss ideas or challenges around multilingual conversational AI and automating customer service in niche industries.

Thanks, HN—we’re eager to answer your questions and hear your thoughts!

Comments URL: https://news.ycombinator.com/item?id=42196517

Points: 1

# Comments: 0

Article URL: https://langfuse.com/changelog/2024-11-20-full-multi-modal-images-audio-attachments

Comments URL: https://news.ycombinator.com/item?id=42196497

Points: 1

# Comments: 0

Article URL: https://lbrito.ca/blog/2023/12/leaving-amazon.html

Comments URL: https://news.ycombinator.com/item?id=42196495

Points: 1

# Comments: 0

Article URL: https://cacm.acm.org/opinion/what-is-a-bug/

Comments URL: https://news.ycombinator.com/item?id=42196488

Points: 1

# Comments: 0

Article URL: https://kinduff.com/2024/11/19/power-to-the-power-users/

Comments URL: https://news.ycombinator.com/item?id=42196465

Points: 1

# Comments: 0

Article URL: https://lexfridman.com/javier-milei/

Comments URL: https://news.ycombinator.com/item?id=42196464

Points: 2

# Comments: 0

Article URL: https://www.theverge.com/2024/11/20/24300547/xbox-cloud-gaming-game-library-streaming

Comments URL: https://news.ycombinator.com/item?id=42196418

Points: 2

# Comments: 0

Article URL: https://www.crunchydata.com/blog/crunchy-data-warehouse-postgres-with-iceberg-for-high-performance-analytics

Comments URL: https://news.ycombinator.com/item?id=42196394

Points: 4

# Comments: 0

Hey HN,

We’re building Weave: an ML-powered tool to measure engineering output, that actually understands engineering output!

Why? Here’s the thing: almost every eng leader already measures output - either openly or behind closed doors. But they rely on metrics like lines of code (correlation with effort: ~0.3), number of PRs, or story points (slightly better at ~0.35). These metrics are, frankly, terrible proxies for productivity.

We’ve developed a custom model that analyzes code and its impact directly, with a far better 0.94 correlation. The result? A standardized engineering output metric that doesn’t reward vanity. Even better, you can benchmark your team’s output against peers while keeping everything private.

Although this one metric is much better than anything else out there, of course it still doesn't tell the whole story. In the future, we’ll build more metrics that go deeper into things like code quality and technical leadership. And we'll build actionable suggestions on top of all of it to help teams improve and track progress.

After testing with several startups, the feedback has been fantastic, so we’re opening it up today. Connect your GitHub and see what WorkWeave can tell you: https://app.workweave.ai/welcome.

I’ll be around all day to chat, answer questions, or take a beating. Fire away!

Comments URL: https://news.ycombinator.com/item?id=42196381

Points: 2

# Comments: 0

I'm looking for AI-based tools that can help me understand a system's legacy components. Ideally, I would provide the source code, and the model would generate text or diagrams that explain the code's underlying business rules as much as possible.

The broad range of existing LLM tooling seems to focus specifically on the code generation part, but I'm not interested in getting generated code. I love to code, so I would like these models to help me in the boring part of understanding legacy code before the modernization.

The only relevant article I have found so far is https://martinfowler.com/articles/legacy-modernization-gen-ai.html . However, it's based on a private tool by Thoughtworks.

Comments URL: https://news.ycombinator.com/item?id=42196370

Points: 1

# Comments: 0

Article URL: https://github.com/argilla-io/argilla

Comments URL: https://news.ycombinator.com/item?id=42196361

Points: 1

# Comments: 0

Article URL: https://grimminck.medium.com/json-smuggling-a-far-fetched-intrusion-detection-evasion-technique-51ed8f5ee05f

Comments URL: https://news.ycombinator.com/item?id=42196360

Points: 1

# Comments: 0

Article URL: https://thehustle.co/originals/the-thriving-business-of-looking-good-on-zoom

Comments URL: https://news.ycombinator.com/item?id=42196355

Points: 1

# Comments: 0

Article URL: https://www.roadtovr.com/quest-3-pairing-windows-11-release-date/

Comments URL: https://news.ycombinator.com/item?id=42196348

Points: 1

# Comments: 0

Article URL: https://www.justice.gov/opa/pr/brothers-residing-new-york-and-bangladesh-charged-running-digital-streaming-piracy-scheme

Comments URL: https://news.ycombinator.com/item?id=42196333

Points: 1

# Comments: 0

Article URL: https://www.justice.gov/opa/pr/california-man-sentenced-acting-illegal-agent-peoples-republic-china-government-and-bribery

Comments URL: https://news.ycombinator.com/item?id=42196321

Points: 3

# Comments: 0

Article URL: https://lws.io/blog/you-got-this-closing/

Comments URL: https://news.ycombinator.com/item?id=42196306

Points: 2

# Comments: 0