Skip to main content
Back

An AI-Based Software Engineer Just Shook The Web

VIEW DEMO

March 14, 2024 4 min read

Hey,

Welcome to this week's edition of AlphaSignal. 


Whether you are a researcher, engineer, developer, or data scientist, our summaries ensure you're always up-to-date with the latest breakthroughs in AI. 


Let's get into it!


Lior

IN TODAY'S EMAIL

  • Devin: The First AI Software Engineer

  • Top News

  • Python tip: Partial Function Application

Read Time: 4 min 01 sec

TOP NEWS

Devin: The First AI Software Engineer

What's New?

Devin is a new model that can complete jobs on Upwork, pass interviews at top AI companies, and excel at the SWE-Bench coding benchmark. Launched by Cognition and backed by industry giants such as Patrick and John Collison, Elad Gil, and Peter Thiel's Founders Fund, Devin launch video reached 27M views on X.

Features:

  • Handles all aspects of software development, from coding to deployment.
  • Operates in a sandboxed environment with access to standard developer tools.
  • Facilitates user interaction through a natural language interface for real-time monitoring and commands.
  • Demonstrates superior performance in SWE-Bench, resolving 13.86% of issues unassisted.

Devin showcases significant advancements in autonomous software problem-solving and project execution. This model's introduction to the market indicates a shift towards more sophisticated AI roles in software development.


Currently, access to Devin is limited to a select group of users, with Cognition planning broader availability in the future. Interested individuals or organizations can reach out for early access through Cognition's contact channels, such as their official website or direct email at info@cognition-labs.com.


The model's underlying technology, attributed to advances in long-term reasoning and planning, remains proprietary. However, its performance in benchmarks and real-world applications highlights the potential for AI to take on more integral roles in software engineering tasks.

View demo

Experience breakthrough multi-workload performance in the cloud.

Get the most out of the L40S by relying on Latitude.sh’s high-performance networking and enterprise-grade hardware designed for Machine Learning engineers. With an intuitive dashboard and pre-installed AI software, you get from zero to running workloads in just a few clicks.


NVIDIA's universal GPU is now available at Latitude.sh in fully-dedicated clusters and container GPUs. With 48 GB of vRAM, the GPU is a great value for universal workloads: from running inference to training models with up to 24 Billion parameters in FP16.

Get started

partner with us

TOP NEWS

Figure partners with OpenAI to give a voice to their robots

---------------------

“With OpenAI, Figure 01 can now have full conversations with people” 


Figure, in collaboration with OpenAI, released Figure 01, a humanoid robot utilizing a multimodal model for real-time interactions and task execution. The model processes visual and textual data, enabling tasks like object identification and garbage collection. “Our robot can describe its visual experience, plan future actions, reflect on its memory, and explain its reasoning verbally,”

Anthropic releases their fastest model yet: Haiku

---------------------
Anthropic released Claude 3 Haiku, an AI model processing 21K tokens/second for sub-32K token prompts, targeting enterprise needs. It utilizes a 1:5 input-to-output token ratio for cost-effective large dataset analysis. Rigorous security protocols ensure enterprise-grade protection. Available through the Claude API and Claude Pro, it efficiently processes extensive documents like Supreme Court cases.

DeepMind announces SIMA, an AI that can perform tasks in 3D Games AI that can perform tasks in 3D Games

---------------------

Google DeepMind has released SIMA, a generalist AI agent capable of executing tasks in diverse 3D game environments based on natural-language instructions. It utilizes pre-trained vision and language models, integrating image-language mapping and action prediction without game-specific coding. Trained across multiple games, SIMA demonstrates the ability to generalize skills and understand instructions, significantly outperforming game-specific agents.

Meta releases new details on GenAI Infrastructure

---------------------

Meta has unveiled two large-scale AI training clusters, each with 24,576 NVIDIA H100 GPUs, to train the Llama 3 models. Built on the Grand Teton open-source platform and using PyTorch, these clusters test RoCE and NVIDIA Quantum2 InfiniBand networks to enhance scalability and performance in AI workloads.

PYTORCH TIP

Partial Function Application

The partial function from the functools module in Python allows you to create a new function by partially applying arguments to an existing function. This can be particularly useful when you need to reuse a function with certain arguments fixed, or when you want to create specialized versions of a function without rewriting it entirely.


Here's an example:


from functools import partial

def my_function(a, b, c):
    return a + b * c

# Create a new function with 'a' fixed to 10
new_function = partial(my_function, 10)

# Call the new function with only 'b' and 'c' arguments
result = new_function(2, 3)
print(result)  # Output: 16

In this example, new_function is a partially applied version of my_function where the first argument a is fixed to 10. When you call new_function with 2 and 3, it effectively calls my_function(10, 2, 3), resulting in the output 16.


Partial function application can be particularly useful in machine learning tasks, where you often need to apply the same function with different sets of arguments or parameters. For example, you could create specialized versions of a model evaluation function with different evaluation metrics or data preprocessing steps partially applied.

Not Great   Good    Amazing

Ghita is an editor at AlphaSignal and AI Engineer at Clear Ventures. Experienced in Computer Vision and Language models, she holds 2 masters in Applied Mathematics.

Thank You.

Want to promote your company, product, job, or event to 175,000+ AI researchers and engineers? You can reach out here.