Cycle 5: Parviainen - Specreduce development

finance/proposal-calls/cycle5/parviainen_specreduce.md

@@ -0,0 +1,89 @@
+### Specreduce Development and Project Management
+
+### Project Team
+
+- Hannu Parviainen
+
+
+### Project Description / Scope of Work
+
+This proposal requests funding to support my continued development and project 
+management work on Specreduce.
+
+The main goal of the proposed work is to ensure that Specreduce provides a 
+complete, reliable, and user-friendly toolset for the reduction of long-slit 
+spectra, supported by comprehensive and approachable documentation, examples, 
+and tutorials. Once this stage is complete, the next phase will extend 
+Specreduce’s capabilities to support multi-object spectrographs.
+
+
+#### Roadmap Items
+
+- :large_orange_diamond: *"Improve integration of core package and coordinated 
+  package documentation from a user perspective."*
+
+    - A major component of this proposal involves improving the clarity and 
+      structure of the Specreduce documentation.  Ongoing work can be followed in 
+      [Specreduce PR #275](https://specreduce--275.org.readthedocs.build/en/275/).
+
+- :large_orange_diamond: *"Develop, host, and index tutorials suitable for use 
+  in university astronomy courses."*
+
+  - The documentation work will include the creation of fully worked tutorials 
+  and teaching materials, such as the [LHIRES-III tutorial](https://github.com/hpparvi/specreduce_example_lhires/blob/main/example_1.ipynb)
+  that can be directly used in teaching spectroscopy in a university setting.
+
+- :red_square: *"Support JIT compilation (e.g., numba, JAX, etc.) throughout 
+   Astropy core and coordinated packages."*
+
+  - The newly developed functionality will be accelerated using Numba where 
+    appropriate, and existing routines can also be optimized (without changing 
+    the API) once all user-visible features are in place. (I have used Numba 
+    extensively in my own packages since 2016 and consider myself fairly
+    familiar with it.)
+
+
+#### Project / Work / Deliverables
+
+- Complete the development of missing or partially implemented specreduce core 
+  functionality, including:
+  - flux calibration,
+  - uncertainty propagation, and
+  - PSF-based spectral extraction.
+
+- Implement the infrastructure required for multi-object spectrograph data 
+  reduction.
+
+- Create a fully defined set of tutorials and educational materials for 
+  spectroscopic data reduction using Specreduce. These materials will be 
+  published as a “Learn Astropy” guide and will include step-by-step *recipes* 
+  covering common spectroscopic reduction tasks for a variety of instruments 
+  and setups.
+
+- Continue Specreduce project management and maintenance, including:
+  - reviewing pull requests and managing GitHub issues,
+  - coordinating community contributions,
+  - preparing and publishing new releases,
+  - maintaining synchronization between PyPI and Conda distributions, and
+  - ensuring alignment with Astropy’s development and contribution guidelines.
+
+
+### Approximate Budget
+
+The proposed optimal budget is based on an **hourly rate of USD 150** and an 
+**anticipated workload of three hours per week over nine months**, from April 
+to December 2026 (the first three months of 2026 are covered by existing 
+funding). This corresponds to a total of 108 hours and an **estimated project 
+cost of USD 16,200.**
+
+The project can also be carried out at a reduced or slightly expanded capacity. 
+The minimum viable budget would support a workload of one hour per week, 
+resulting in a total cost of USD 5,400, while a modestly enhanced workload of 
+four hours per week would bring the total to USD 21,600. Although one hour 
+per week would significantly slow the development of new functionality, it 
+would still allow for responsive project management and maintenance.
+
+
+### Period of Performance
+
+April 1, 2026 – December 31, 2026