Coding & Dev AI Tools

Portkey

AI gateway that unifies LLM providers with reliability and observability

Portkey is an AI gateway that routes requests across 100+ LLM providers with features like load balancing, fallbacks, caching, and rate limiting. It also provides a full observability suite with request logging, cost tracking, and prompt versioning. Built for teams that run LLMs in production and need resilience and cost control.

Amazon CodeWhisperer

Amazon's AI code companion for secure, cloud-aware code generation

Amazon CodeWhisperer is an AI coding assistant from AWS that generates real-time code suggestions and completions directly in popular IDEs including VS Code, JetBrains, and AWS Cloud9. It is trained on Amazon's internal codebase and public code, with built-in security scanning to flag vulnerabilities. Best for developers building on AWS who want code suggestions tuned to AWS APIs and best practices.

Haystack

Open-source framework for building LLM-powered search pipelines

Haystack is an open-source Python framework by deepset for building production-ready NLP applications, including RAG pipelines, semantic search, and question-answering systems. It supports a wide range of LLMs and document stores, giving ML engineers and developers the composable building blocks to ship robust AI applications.

Agenta

Open-source LLMOps platform for prompt engineering and evaluation

Agenta is an open-source platform for building, versioning, and evaluating LLM applications with a focus on prompt engineering workflows. It provides a playground for testing prompts, A/B comparison, human evaluation tools, and deployment pipelines. Suited for AI engineers and product teams who need a structured approach to iterating on LLM-powered features.

Amazon Q Developer CLI

AI-powered CLI assistant for AWS development and cloud tasks

Amazon Q Developer CLI brings AWS's generative AI assistant directly into the terminal, enabling developers to get code suggestions, explain shell commands, and automate AWS tasks through natural language. It integrates with the AWS CLI and supports workflows like infrastructure setup, debugging, and service configuration. Built for developers and cloud engineers working in AWS environments.

LlamaIndex

The leading data framework for building LLM-powered applications

LlamaIndex is an open-source data framework that connects custom data sources to large language models for building retrieval-augmented generation (RAG) pipelines and AI agents. It supports ingesting, indexing, and querying documents, databases, and APIs with dozens of integrations. Widely used by developers and enterprises building production LLM applications.

CodeRabbit

AI-powered code review that provides instant, contextual PR feedback

CodeRabbit is an AI code review tool that automatically analyzes pull requests and provides line-by-line feedback on bugs, security issues, style violations, and logic errors. It integrates with GitHub and GitLab and learns from reviewer feedback over time. Suitable for engineering teams who want to accelerate code review and improve code quality.

Kilo Code

Open-source AI coding agent for VS Code with deep codebase context

Kilo Code is an open-source AI coding agent extension for VS Code that provides code generation, refactoring, and multi-file editing with full project context awareness. It supports multiple LLM backends and lets developers bring their own API keys for full control. Aimed at developers who want a powerful, customizable AI coding assistant without vendor lock-in.

tabnine

Privacy-first AI code completion that runs locally or in the cloud

Tabnine is an AI code completion tool that provides whole-line and full-function suggestions across all major languages and IDEs. It can run fully on-premise or locally to keep proprietary code private, making it a popular choice for enterprises with strict data governance requirements. Supports VS Code, JetBrains, Vim, and more with team-specific model personalization.

Debuild

AI-powered low-code tool for building web apps from descriptions

Debuild is an AI-assisted low-code web application builder that generates React components and SQL queries from plain text descriptions. Users describe what they want and Debuild assembles the frontend and backend logic automatically. Geared toward developers and entrepreneurs who want to prototype and launch web apps with minimal manual coding.

Callstack.ai PR Reviewer

AI code reviewer specialized for React Native and JavaScript PRs

Callstack.ai PR Reviewer is an AI-powered pull request review tool built by the React Native experts at Callstack. It provides automated, contextual feedback on JavaScript and React Native code, covering performance, best practices, and React-specific patterns. Tailored for React Native teams who want domain-specific AI review beyond generic linting.

Phoenix

Open-source AI observability platform for evaluating LLM apps

Phoenix by Arize is an open-source observability and evaluation tool for AI and LLM applications. It provides tracing, dataset management, and built-in evaluators to measure hallucination, relevance, and toxicity. Designed for ML engineers and data scientists who need deep insight into model and agent behavior during development and production.

LLM

Run and log prompts to any LLM from the command line

LLM is a CLI tool and Python library by Simon Willison (Datasette) for running prompts against OpenAI, Anthropic, Gemini, and local models from the terminal. It logs every prompt and response to a local SQLite database and supports plugins for additional model providers. Ideal for developers and researchers who want scriptable, auditable LLM access.

Gopher

DeepMind's 280B parameter language model for research

Gopher is a 280-billion parameter large language model developed by DeepMind, designed to advance understanding of language model capabilities and limitations at scale. It was evaluated across 152 diverse tasks and demonstrated strong performance in reading comprehension, fact-checking, and knowledge-intensive domains. Released as a research model, it also included a companion paper on ethical considerations and societal impact.

OpenAI Downtime Monitor

Real-time status tracker for OpenAI API uptime and incidents

OpenAI Downtime Monitor, powered by Portkey, tracks the live availability and latency of OpenAI's API and related services. It shows historical uptime, active incidents, and response time trends across model endpoints. Useful for developers and teams who depend on OpenAI APIs and need instant visibility into service disruptions.

GitHub Copilot

AI pair programmer that writes code alongside you in real time

GitHub Copilot is an AI-powered coding assistant developed by GitHub and OpenAI, integrated directly into editors like VS Code, JetBrains, and Neovim. It suggests code completions, generates entire functions, writes tests, and explains code based on context. Widely used by developers to accelerate software development and reduce boilerplate.

SEAL LLM Leaderboard

Scale AI's expert-evaluated leaderboard for LLM capabilities

The SEAL (Scale Evaluation and Assessment of Language Models) Leaderboard by Scale AI provides rigorous, expert-human-evaluated rankings of LLMs across domains like coding, instruction following, math, and reasoning. Unlike automated benchmarks, SEAL uses human raters to assess model quality. It is a trusted reference for enterprises and AI researchers selecting models for demanding applications.

Codeflash

AI tool that automatically optimizes Python code for performance

Codeflash is an AI-powered Python performance optimization tool that automatically finds and rewrites slow code to make it faster. It generates optimized versions of functions, validates correctness with tests, and benchmarks improvements before suggesting changes. Built for Python developers and data engineers who want measurable performance gains without manual profiling.

Together AI

Run and fine-tune leading open-source LLMs at cloud scale

Together AI is a cloud platform for deploying, fine-tuning, and serving open-source foundation models such as Llama, Mistral, and FLUX at high throughput. It offers a unified API compatible with OpenAI's interface and dedicated GPU instances. Ideal for AI teams building production applications who need cost-efficient, scalable inference.

Opik

Open-source LLM evaluation and tracing platform by Comet

Opik is an open-source LLM evaluation and observability tool from Comet that helps teams trace, evaluate, and debug AI applications. It provides tracing of LLM calls, dataset management, and built-in and custom evaluators for measuring output quality. Designed for ML engineers who need a production-ready evaluation pipeline integrated with their experiment tracking.

DeepSeek

China's frontier open-source LLM rivaling the world's best

DeepSeek is a Chinese AI lab offering a series of highly capable open-source large language models, including DeepSeek-V2 and DeepSeek-R1, which rival proprietary models on reasoning and coding benchmarks. The models are available via API and for self-hosting. DeepSeek attracted global attention for delivering frontier performance at low cost.

Bloom

Open multilingual LLM trained by 1,000+ researchers worldwide

BLOOM is a 176-billion parameter open-access multilingual language model developed collaboratively by over 1,000 AI researchers through the BigScience project. It supports 46 natural languages and 13 programming languages, making it one of the most linguistically diverse LLMs available. BLOOM is freely available on Hugging Face and is designed for researchers and developers needing a transparent, open-source alternative to proprietary models.

OPT

Meta's open pretrained transformer language models for researchers

OPT (Open Pretrained Transformers) is a suite of open-source large language models from Meta AI, ranging from 125M to 175B parameters, designed as a reproducible and accessible alternative to GPT-3. The models are available on Hugging Face for research purposes and are trained on publicly available datasets. OPT aims to enable the broader research community to study large language model behavior and limitations.

v0

Generate production-ready UI components from a text prompt

v0 by Vercel is a generative UI tool that converts natural language prompts into React and Tailwind CSS components ready to ship. It supports iterative refinement through chat, exports clean code, and integrates with Next.js projects. Perfect for developers and designers who want to scaffold polished frontends in seconds.

Stable Beluga

Stability AI's instruction-tuned 65B LLM built on LLaMA

Stable Beluga is a 65-billion parameter language model from Stability AI, fine-tuned on LLaMA 65B using an internal Orca-style dataset to enhance instruction following. It excels at answering questions, completing writing tasks, and following complex instructions. Released under a non-commercial CC BY-NC-4.0 license, it targets researchers and developers building open-source LLM applications.

LangChain

The leading framework for building LLM-powered applications

LangChain is an open-source Python and JavaScript framework that provides the core abstractions and integrations for building applications powered by large language models, including agents, RAG pipelines, and tool-using systems. It is the most widely adopted LLM application framework, used by developers ranging from hobbyists to Fortune 500 engineering teams.

Build an AI Agent (From Scratch)

Practical guide to designing and coding autonomous AI agents

This Manning Publications book provides a step-by-step approach to building autonomous AI agents that can plan, use tools, browse the web, and execute multi-step tasks. It covers agentic frameworks, memory systems, tool calling, and evaluation strategies with hands-on Python implementations. The book is aimed at developers and AI engineers who want to move from passive LLM usage to building production-ready agentic systems.

Build a Large Language Model (From Scratch)

Learn to build an LLM from the ground up with hands-on code

This Manning Publications book by Sebastian Raschka guides readers through building a GPT-style large language model from scratch using Python and PyTorch. It covers tokenization, transformer architecture, pre-training, and fine-tuning with clear, step-by-step code examples. The book is designed for ML practitioners and engineers who want a deep, practical understanding of how modern LLMs like ChatGPT actually work under the hood.

Context Engineering Guide: Prompt Engineering Guide

In-depth guide to context engineering for LLM applications

The Context Engineering Guide from PromptingGuide.ai provides a thorough overview of techniques for designing and managing the information provided to language models beyond simple prompt writing. It covers retrieval-augmented generation, context window optimization, structured context schemas, and evaluation methods for context quality. The guide is valuable for AI engineers and researchers building sophisticated LLM pipelines who need to go beyond basic prompting.

Google AI Studio

Prototype and test Gemini models with a free browser interface

Google AI Studio is a free, browser-based developer environment for testing and building with the Gemini family of models, including Gemini 1.5 Pro's million-token context. It supports multimodal inputs, text, images, audio, video, and generates API keys and code snippets for fast integration. Ideal for developers exploring Gemini's capabilities before building full applications.

ChatGPT prompt engineering for developers

DeepLearning.AI short course on prompt engineering with GPT

A free short course by DeepLearning.AI and OpenAI, taught by Isa Fulford and Andrew Ng, covering best practices for prompting ChatGPT in application development. Topics include iterative prompt refinement, summarization, inference, transformation, and building chatbots. Ideal for developers integrating LLMs into software products.

Myriad

Generate diverse AI content prompts to spark creative ideas fast

Myriad is a free AI content prompt tool by NamePepper that helps writers, marketers, and creators generate varied prompt ideas for blog posts, social content, and campaigns. It quickly surfaces angles and hooks so teams can beat creative blocks and produce content more efficiently.

Copilot

Microsoft's AI companion powered by GPT-4 across all products

Microsoft Copilot is an AI assistant integrated across Microsoft's ecosystem including Windows, Edge, Microsoft 365, and Bing. Powered by OpenAI's GPT-4, it helps with writing, research, image generation, coding, and task automation. Available for free with optional paid plans for enhanced productivity features.

Gemini CLI

Google's open-source AI agent for coding tasks in the terminal

Gemini CLI is an open-source command-line AI agent by Google that uses Gemini models to assist with coding, file editing, and agentic tasks directly in the terminal. It connects to Google Search for grounding and supports tool use for real-world actions. Aimed at developers who want a capable, free, open-source terminal AI assistant.

Build a Reasoning Model (From Scratch)

Hands-on guide to building reasoning-capable AI models step by step

This Manning Publications book teaches readers how to construct reasoning-focused AI models inspired by systems like OpenAI's o1, covering chain-of-thought training, reinforcement learning from human feedback, and search-augmented inference. It provides practical code implementations alongside the theory to help readers understand how deliberative reasoning emerges in modern LLMs. The book is suited for ML engineers and researchers looking to go beyond standard instruction fine-tuning.

Calmo

AI-powered customer support automation for lean support teams

Calmo is an AI customer support tool that automates responses to common inquiries using your existing knowledge base and documentation. It handles ticket deflection, drafts replies for agent review, and escalates complex issues to humans. Designed for small and mid-sized teams looking to scale support quality without growing headcount.