Merge branch 'main' into issue_242

Author: Kushagra7777

.gitignore

@@ -164,3 +164,6 @@ cython_debug/
 
 # VS Code
 .vscode/
+
+# logs
+lightning_logs/

Makefile

@@ -6,4 +6,4 @@ serve-docs:
 .PHONY: deploy-docs
 deploy-docs:
 	uv run --group docs mkdocs build
-	uv run --group docs modal deploy .github/deploy_docs.py
+	uv run --group docs modal deploy .github/deploy_docs.py

README.md

@@ -6,10 +6,10 @@
   <em>The GenAI Forecasting Agent · LLMs × Time Series Foundation Models</em>
 </div>
 <div align="center">
-  <a href="https://github.com/AzulGarza/TimeCopilot/actions/workflows/ci.yaml"><img src="https://github.com/AzulGarza/TimeCopilot/actions/workflows/ci.yaml/badge.svg?branch=main" alt="CI"></a>
+  <a href="https://github.com/TimeCopilot/timecopilot/actions/workflows/ci.yaml"><img src="https://github.com/TimeCopilot/timecopilot/actions/workflows/ci.yaml/badge.svg?branch=main" alt="CI"></a>
   <a href="https://pypi.python.org/pypi/timecopilot"><img src="https://img.shields.io/pypi/v/timecopilot.svg" alt="PyPI"></a>
-  <a href="https://github.com/AzulGarza/timecopilot"><img src="https://img.shields.io/pypi/pyversions/timecopilot.svg" alt="versions"></a>
-  <a href="https://github.com/AzulGarza/timecopilot/blob/main/LICENSE"><img src="https://img.shields.io/github/license/azulgarza/timecopilot.svg" alt="license"></a>
+  <a href="https://github.com/TimeCopilot/timecopilot"><img src="https://img.shields.io/pypi/pyversions/timecopilot.svg" alt="versions"></a>
+  <a href="https://github.com/TimeCopilot/timecopilot/blob/main/LICENSE"><img src="https://img.shields.io/github/license/TimeCopilot/timecopilot.svg" alt="license"></a>
   <a href="https://discord.gg/7GEdHR6Pfg"><img src="https://img.shields.io/discord/1387291858513821776?label=discord" alt="Join Discord" /></a>
 </div>
 
@@ -26,6 +26,20 @@ Developed with 💙 at [timecopilot.dev](https://timecopilot.dev/).
 
 ---
 
+## 🚀 Key Capabilities
+
+- **Unified Forecasting Layer**. Combines 30+ time-series foundation models (Chronos, Moirai, TimesFM, TimeGPT…) with LLM reasoning for automated model selection and explanation.
+
+- **Natural-Language Forecasting**. Ask questions in plain English and get forecasts, analysis, validation, and model comparisons. No scripts, pipelines, or dashboards needed.
+
+- **One-Line Forecasting**. Run end-to-end forecasts on any dataset in seconds with a single command (`uvx timecopilot forecast <url>`).
+
+## 📰 News
+
+- **#1 on the GIFT-Eval benchmark**. TimeCopilot reached the [top position](https://timecopilot.dev/experiments/gift-eval/) on the global GIFT-Eval benchmark above AWS, Salesforce, Google, IBM, and top universities.  
+- **Accepted at NeurIPS (BERTs Workshop)**. Our [work on agentic forecasting](https://arxiv.org/pdf/2509.00616) was [accepted](https://berts-workshop.github.io/accepted-papers/) at NeurIPS 2025.
+- **New Manifesto: “Forecasting, the Agentic Way”.**  Our [founding essay](https://timecopilot.dev/blog/forecasting-the-agentic-way/) on agentic forecasting, TSFMs, and the future of time series.
+
 ## How It Works
 
 TimeCopilot is a generative agent that applies a systematic forecasting approach using large language models (LLMs) to:

docs/blog/.authors.yml

@@ -0,0 +1,11 @@
+authors:
+  azulgarza:
+    name: Azul Garza 
+    description: Co-Creator
+    avatar: https://avatars.githubusercontent.com/u/10517170
+    url: https://github.com/AzulGarza
+  reneerosillo:
+    name: Renee Rosillo
+    description: Co-Creator
+    avatar: https://media.licdn.com/dms/image/v2/D4E03AQGN9Z-UddZReg/profile-displayphoto-shrink_800_800/profile-displayphoto-shrink_800_800/0/1727908622546?e=1766016000&v=beta&t=zDLuE5S7jSlE_xDfQwPSTJnZjIi-BL_fzO44f8VnMkg
+    url: https://www.linkedin.com/in/reneerosillo/

docs/blog/.meta.yml

@@ -0,0 +1,2 @@
+hide:
+  - feedback

docs/blog/index.md

@@ -0,0 +1 @@
+# Blog

docs/blog/posts/forecasting-the-agentic-way.md

@@ -0,0 +1,50 @@
+---
+date: 2025-12-01
+authors:
+  - azulgarza
+  - reneerosillo
+categories:
+  - General
+description: >
+  We are thrilled to announce TimeCopilot
+title: Forecasting, the Agentic Way 
+slug: forecasting-the-agentic-way
+---
+
+
+# Forecasting, the Agentic Way
+
+TimeCopilot is open-source agentic forecasting with LLMs and the strongest time series models. We started TimeCopilot with a clear goal. We want to democratize time series forecasting and make it accessible not only to humans but to the next wave of automated agentic systems. Accurate, automated, and easy forecasts.
+
+## We are all grounded by temporal data
+
+In 2005, Steve Jobs [told the Stanford graduating class](https://www.youtube.com/watch?v=UF8uR6Z6KLc) that we can only connect the dots looking backwards. At a fundamental level, he named one of the forces that shapes how we live. He explained how we move through the world without knowing what our actions will create, and why trusting those actions matters when the outcome is uncertain. For professionals who work with forecasts every day, this is not an abstract idea. Our work supports planning, risk, demand, supply, energy, finance, operations, and more. And as Steve said, the success of our work depends on how well we use past outcomes in the form of time series data to see what the future might look like through the lens of our own context.
+
+Daily forecasting work is not glamorous. It is the nitty-gritty of managing pipelines, models, baselines, assumptions, overrides, and schedules. It means moving between scripts, dashboards, notebooks, and model families. Many forecasters know that explaining a forecast often takes more time than producing one. 
+
+Why go to so much trouble to understand the future? As a field we know that [small shifts in starting conditions can create very different outcomes](https://en.wikipedia.org/wiki/Lorenz_system). And we know that uncertainty has structure. When we understand that structure, we make better decisions. There is satisfaction in taking something complex and turning it into a pattern we can recognize. In simple terms, we do it because we enjoy predicting the future. TimeCopilot is our rendition to the field. It reflects both technical progress and an ideal. We want to use all the technology we have to make better predictions.
+
+## We believe the future opens up with LLMs
+
+Forecasting has moved through several eras. Classical statistics. Probabilistic approaches. Deep learning. And now time series foundation models trained on large and diverse datasets. Each method brings value, but none solves the full problem. Different inputs and assumptions need different tools. Every forecaster knows: [no single model, no matter how powerful, can solve every problem](https://en.wikipedia.org/wiki/No_free_lunch_theorem). Forecasting fuel needs for a high tide water platform is not the same as forecasting train schedules for a small province in Germany. Our work as forecasters has always relied on understanding context and resetting the constraints that make specific pipelines work for specific use cases. But what if we had an automated helper that understood our particular needs? With the rise of agentic systems, we finally have an extra set of hands to take on the repetitive work we have to deal with every day. 
+
+As we see it, there’s a need to build a layer above all the foundation models and the systems that came before. A layer that knows when to use each method, how to combine them, and how to explain the reasoning behind a forecast.
+
+TimeCopilot is that layer. It uses a wide set of time series foundation models and classical models and libraries. It applies LLMs to add context, constraints, and reasoning. It explains its assumptions in natural language. It updates forecasts as new information arrives. It lets people work through text while keeping the granularity that teams need of hand-tuning pipelines: a coordinated reasoning layer that uses many tools and adapts as context shifts.
+
+With the dawn of large language models, it became evident that written word matters, and plain English may be the most natural interface for reasoning. Andrej Karpathy often notes that [language is becoming the new interface for software](https://x.com/karpathy/status/1617979122625712128?lang=en). Forecasting fits that idea. The work is built on uncertainty, intent, and assumptions. Text gives us a simple way to express that. Text translates our needs and helps us define the guardrails for our context. TimeCopilot is the agent for forecasting in the same way [Clay](https://www.clay.com/) helps marketing teams move faster and [Cursor](https://cursor.com/) helps engineering teams ship code with less friction.
+
+## And so we asked ourselves a simple question: can we use agents for time series forecasting?
+
+TimeCopilot began as a first principles question about forecasting and context. Early prototypes led to a research effort. Our early project attracted collaborators, engineers, and contributors who cared about the same ideas. In a short time, it has become a meaningful open source time series project. We have assembled [the largest unified collection](https://timecopilot.dev/model-hub/) of time series foundation models with more than 30 models across more than 7 families. We have shown practical examples of [agentic forecasting](https://timecopilot.dev/examples/llm-providers/). We reached the top position on the [GIFT Eval benchmark](https://timecopilot.dev/experiments/gift-eval/) above AWS, Salesforce, IBM, and top universities. We [published at NeurIPS](https://berts-workshop.github.io/accepted-papers/). We passed 12k downloads, 280 GitHub stars, and continue to grow a community around this work. Everything is open source and built with a scientific mindset.
+
+We now have early support that helps us move quickly and work with organizations exploring where agentic forecasting can be useful.
+
+We believe the next generation of forecasting will be grounded in data, expressed through text, orchestrated across many models, clear about uncertainty, and built in the open.
+
+TimeCopilot is our next step towards that. 
+
+More soon.
+
+Azul & Renée
+

docs/changelogs/v0.0.10.md

@@ -1,6 +1,6 @@
 
 ### Features
-* **TimeCopilotForecaster Class**: Introduced the `TimeCopilotForecaster` class to enhance forecasting capabilities. See [#48](https://github.com/AzulGarza/timecopilot/pull/48). Example:
+* **TimeCopilotForecaster Class**: Introduced the `TimeCopilotForecaster` class to enhance forecasting capabilities. See [#48](https://github.com/TimeCopilot/timecopilot/pull/48). Example:
 
 ```python
 import pandas as pd
@@ -17,7 +17,7 @@ forecaster = TimeCopilotForecaster(models=[TimesFM(), SeasonalNaive()])
 fcsts_df = forecaster.forecast(df=df, h=12, freq="MS")
 ```
 
-* **Probabilistic Forecasts**: Added support for probabilistic forecasts in the forecaster class. See [#50](https://github.com/AzulGarza/timecopilot/pull/50). Example:
+* **Probabilistic Forecasts**: Added support for probabilistic forecasts in the forecaster class. See [#50](https://github.com/TimeCopilot/timecopilot/pull/50). Example:
 
 ```python
 import pandas as pd
@@ -46,12 +46,12 @@ fcsts_df_quantiles = forecaster.forecast(
 ```
 
 * **Integration with External Libraries**:
-    - **timesfm**: Added Google's foundation model [TimesFM](https://github.com/google-research/timesfm). See [#55](https://github.com/AzulGarza/timecopilot/pull/55).
-    - **chronos**: Added AWS AI Labs's foundation model [Chronos](https://arxiv.org/abs/2403.07815). See [#59](https://github.com/AzulGarza/timecopilot/pull/59).
-    - **Prophet**: Added Facebook's [Prophet](https://facebook.github.io/prophet/) to available models. See [#61](https://github.com/AzulGarza/timecopilot/pull/61).
+    - **timesfm**: Added Google's foundation model [TimesFM](https://github.com/google-research/timesfm). See [#55](https://github.com/TimeCopilot/timecopilot/pull/55).
+    - **chronos**: Added AWS AI Labs's foundation model [Chronos](https://arxiv.org/abs/2403.07815). See [#59](https://github.com/TimeCopilot/timecopilot/pull/59).
+    - **Prophet**: Added Facebook's [Prophet](https://facebook.github.io/prophet/) to available models. See [#61](https://github.com/TimeCopilot/timecopilot/pull/61).
 
 
-* **Multi-series Support**: Enhanced the agent to handle multiple time series. See [#64](https://github.com/AzulGarza/timecopilot/pull/64).
+* **Multi-series Support**: Enhanced the agent to handle multiple time series. See [#64](https://github.com/TimeCopilot/timecopilot/pull/64).
     - **Example**:
         ```python
         from timecopilot import TimeCopilot
@@ -61,10 +61,10 @@ fcsts_df_quantiles = forecaster.forecast(
         tc.forecast(...)
         ```
 
-* **Agent Integration**: Utilized the TimeCopilotForecaster class within the agent. See [#65](https://github.com/AzulGarza/timecopilot/pull/65).
+* **Agent Integration**: Utilized the TimeCopilotForecaster class within the agent. See [#65](https://github.com/TimeCopilot/timecopilot/pull/65).
 
 ### Tests
-* **Basic Functionality Tests**: Added tests for basic functionality to ensure reliability. See [#43](https://github.com/AzulGarza/timecopilot/pull/43).
+* **Basic Functionality Tests**: Added tests for basic functionality to ensure reliability. See [#43](https://github.com/TimeCopilot/timecopilot/pull/43).
 
 ### Fixes
-* **CI Improvements**: Implemented a fix to cancel concurrent CI runs, optimizing the CI process. See [#63](https://github.com/AzulGarza/timecopilot/pull/63).
+* **CI Improvements**: Implemented a fix to cancel concurrent CI runs, optimizing the CI process. See [#63](https://github.com/TimeCopilot/timecopilot/pull/63).

docs/changelogs/v0.0.11.md

@@ -1,6 +1,6 @@
 ### Features
 
-* **TiRex Foundation Model**: Added the [TiRex](https://github.com/NX-AI/tirex) time series foundation model. See [#77](https://github.com/AzulGarza/timecopilot/pull/77). Example:
+* **TiRex Foundation Model**: Added the [TiRex](https://github.com/NX-AI/tirex) time series foundation model. See [#77](https://github.com/TimeCopilot/timecopilot/pull/77). Example:
 
 ```python
 import pandas as pd
@@ -14,7 +14,7 @@ model = TiRex()
 fcst = model.forecast(df, h=12)
 ```
 
-* **Toto Model**: Added the [Toto](https://github.com/DataDog/toto) time series model. See [#78](https://github.com/AzulGarza/timecopilot/pull/78). Example:
+* **Toto Model**: Added the [Toto](https://github.com/DataDog/toto) time series model. See [#78](https://github.com/TimeCopilot/timecopilot/pull/78). Example:
 
 ```python
 import pandas as pd
@@ -28,7 +28,7 @@ model = Toto()
 fcst = model.forecast(df, h=12)
 ```
 
-* **Optional `freq` Parameter**: The `freq` parameter is now optional in all forecast and cross-validation methods. If not provided, frequency is inferred automatically from the data. See [#96](https://github.com/AzulGarza/timecopilot/pull/96). Example:
+* **Optional `freq` Parameter**: The `freq` parameter is now optional in all forecast and cross-validation methods. If not provided, frequency is inferred automatically from the data. See [#96](https://github.com/TimeCopilot/timecopilot/pull/96). Example:
 
 ```python
 import pandas as pd
@@ -43,11 +43,11 @@ model = SeasonalNaive()
 fcst = model.forecast(df, h=12)
 ```
 
-* **Improved Model Docstrings**: All foundational and statistical model constructors are now fully documented, with clear parameter explanations and references to official sources. See [#93](https://github.com/AzulGarza/timecopilot/pull/93) and [#94](https://github.com/AzulGarza/timecopilot/pull/94).
+* **Improved Model Docstrings**: All foundational and statistical model constructors are now fully documented, with clear parameter explanations and references to official sources. See [#93](https://github.com/TimeCopilot/timecopilot/pull/93) and [#94](https://github.com/TimeCopilot/timecopilot/pull/94).
 
-* **Comprehensive Module Docstrings**: Added module-level docstrings to improve API documentation and usability. See [#82](https://github.com/AzulGarza/timecopilot/pull/82).
+* **Comprehensive Module Docstrings**: Added module-level docstrings to improve API documentation and usability. See [#82](https://github.com/TimeCopilot/timecopilot/pull/82).
 
-* **TimeCopilotForecaster Documentation**: Documented the `TimeCopilotForecaster` class, including its constructor and methods, to clarify its unified, multi-model forecasting and cross-validation interface. See [#97](https://github.com/AzulGarza/timecopilot/pull/97). Example:
+* **TimeCopilotForecaster Documentation**: Documented the `TimeCopilotForecaster` class, including its constructor and methods, to clarify its unified, multi-model forecasting and cross-validation interface. See [#97](https://github.com/TimeCopilot/timecopilot/pull/97). Example:
 
 
 ```python
@@ -75,15 +75,15 @@ cv_df = tcf.cross_validation(df=df, h=12)
 ```
 
 ### Tests
-* **Parallel Test Execution**: Added `pytest-xdist` to enable running tests in parallel, speeding up CI and local test runs. See [#75](https://github.com/AzulGarza/timecopilot/pull/75).
+* **Parallel Test Execution**: Added `pytest-xdist` to enable running tests in parallel, speeding up CI and local test runs. See [#75](https://github.com/TimeCopilot/timecopilot/pull/75).
 
 ### Fixes
 
-* **Documentation Improvements**: Enhanced documentation for all models and constructors, ensuring clarity and consistency across the codebase. See [#93](https://github.com/AzulGarza/timecopilot/pull/93), [#94](https://github.com/AzulGarza/timecopilot/pull/94), and [#82](https://github.com/AzulGarza/timecopilot/pull/82).
+* **Documentation Improvements**: Enhanced documentation for all models and constructors, ensuring clarity and consistency across the codebase. See [#93](https://github.com/TimeCopilot/timecopilot/pull/93), [#94](https://github.com/TimeCopilot/timecopilot/pull/94), and [#82](https://github.com/TimeCopilot/timecopilot/pull/82).
 
-* **S3 Data Source**: All example and test data now use S3 URLs for consistency and reproducibility. See [#73](https://github.com/AzulGarza/timecopilot/pull/73).
+* **S3 Data Source**: All example and test data now use S3 URLs for consistency and reproducibility. See [#73](https://github.com/TimeCopilot/timecopilot/pull/73).
 
 
 
 
-**Full Changelog**: https://github.com/AzulGarza/timecopilot/compare/v0.0.10...v0.0.11
+**Full Changelog**: https://github.com/TimeCopilot/timecopilot/compare/v0.0.10...v0.0.11