JioHotstar Benchmarks Knowledge Bases for AI Coding Agents
Source: InfoQ ·
JioHotstar evaluated 6 approaches to code retrieval for AI coding agents using SWE, CodeScale Bench, and an internal question-answering harness. Structural search performed well on concrete tasks, while semantic retrieval helped with ambiguous requirements. The findings highlight trade-offs in accuracy, token use, and tool design, underscoring the need to match retrieval tools to task types. By Leela Kumili
Frequently Asked Questions
What is JioHotstar Benchmarks Knowledge Bases for AI Coding Agents?
JioHotstar Benchmarks Knowledge Bases for AI Coding Agents is a article discovery curated from InfoQ. JioHotstar evaluated 6 approaches to code retrieval for AI coding agents using SWE, CodeScale Bench, and an internal question-answering harness. Structural search performed well…
How can I access or experience this?
You can experience this content directly on its original host (InfoQ) by clicking the "Open Original" button.
Where does this content come from?
Curated from InfoQ by the OnSuggest team to surface the most interesting and serendipitous content on the web.
More like this
All Technology discoveries →Google’s Tabletop lets groups build AI planning tools together
Google Labs has launched Tabletop, an experiment that builds custom tools such as trip itineraries, dinner polls and expense splitters from a prompt. Groups share one space and edit it together in real time. It is open to US adults via a waitlist. Google has launched Tabletop, an experiment from its Google Labs team that […] This story continues at The Next Web
Helsing launches Clarity, a European rival to Palantir’s Maven
German defence company Helsing has launched Clarity, a direct challenge to Palantir. The AI platform helps armed forces turn drone, satellite and sensor data into targets. Helsing says it builds the software in Europe, under European control. Clarity pulls in sensor feeds, open sources, written reports and the systems a unit already runs. It finds […] This story continues at The Next Web
IBM plans to link two quantum chips into a 1,040-qubit system in 2027
IBM’s next quantum chip, Nighthawk r3, is designed to run 10,000 operations on 520 qubits. IBM plans to link two of them into a 1,040-qubit system in 2027. Customers report speedups of up to 19x on the current Nighthawk chip. IBM plans to build a quantum computer with more than 1,000 qubits in 2027 by […] This story continues at The Next Web
Microsoft turns on Claude by default in UK Copilot, The Register reports
Microsoft will switch on Anthropic’s Claude models by default for UK Copilot customers from 9 December, The Register reported. Anthropic processes that data in the US. Richard Speed wrote for The Register that UK organisations with no setting chosen by then will have Claude enabled. Administrators can still turn it off. Leaving the setting untouched […] This story continues at The Next Web
ElevenLabs moves beyond voice with custom models and a work app
ElevenLabs will help businesses fine-tune AI models on data from their own customer conversations, and opened early access to ElevenAI, an app that takes meeting notes. It also launched a partner programme and marketplace, and expanded its deal with Nvidia. ElevenLabs will help businesses fine-tune AI models on data from their own customer conversations, the […] This story continues at The Next Web
London’s Fractile raises $720M for memory-first AI chips and hits $7B valuation
London-based chip designer Fractile has raised $720 million in a Series C round, betting that memory bandwidth, rather than raw computing power, is the main constraint holding back AI performance. The funding values the four-year-old company at $7 billion, according to the Financial Times, which cited people familiar with the deal. Thrive Capital, Founders Fund, […] This story continues at The Next Web