Hi Victor, James, and Aman, Thanks to all of you for the feedback.
To close this out, the consensus is that revolving credit belongs in a plug-in outside core, is part of the longer-term roadmap, and would require a mentor with functional expertise before it could be considered as a GSoC project. With that in mind, I'll set this idea aside and won't pursue it for GSoC 2027. Regarding James's point about custom modules, one architectural gap worth considering for future work is the lack of an extension point for introducing entirely new product types. Today, modules can add COB business steps, batch jobs, loan transaction processors, and database migrations, but extending Fineract with new product types is not currently supported. This would likely be one of the first areas an FSIP would need to address. I'll focus my efforts on security and stability work instead, in line with James's suggestion, and start a separate thread once I have a concrete proposal. Thanks, Abhishek On Tue, 3 Nov 2026 at 11:31, James Dailey <[email protected]> wrote: > Inline > > On Mon, Nov 2, 2026 at 9:46 PM Abhishek Chaudhary < > [email protected]> wrote: > >> Hi Aman, >> >> Thanks for the detailed reply. >> >> Regarding the existing functionality, Apache Fineract's core does not >> currently have support for a revolving line of credit. The Working Capital >> Loan is the closest existing product, and although its documentation >> describes it as revolving, it is essentially a zero-interest, discount-fee >> loan with sales-based repayments. Its current business logic also permits >> only a single disbursement. There is no credit limit or redraw >> functionality. >> > yes. Indeed. I'm glad you are aware of these functionality differences. > >> I mentioned it as a design reference rather than something to build upon. >> >> I understand and agree with your and James's point. Without a mentor who >> can provide guidance from the functional side, this may not be a suitable >> fit for GSoC 2027. >> >> Unless someone with the relevant functional expertise is interested in >> mentoring the project, I'll set it aside as a GSoC proposal and treat the >> FSIP as a separate, longer-term initiative. >> > Something to consider for the long term roadmap > >> I believe questions around client linkage, RBAC, security, and the >> overall functional design are important aspects that the FSIP should >> address before any implementation work begins. >> > To some extent the "custom module" concept might handle some of these > design elements. > > >> Thanks, >> Abhishek >> >> On Tue, 3 Nov 2026 at 10:09, Aman Mittal <[email protected]> >> wrote: >> >>> Hi Abhishek, >>> >>> Thanks for discussion, i think this functionality should not be the part >>> of the GSOC 2027 due to following reasons. One is that it requires a lot of >>> edge cases (e.g., new custom modules?) how should it be handled? Because it >>> requires someone who can mentor regarding use case perspective (Someone who >>> can guide based on functional requirements). and separate module. I do not >>> know if anyone willing to volunteer in terms of mentoring and then proof of >>> concept. (Already have new repos now for consumer facing and loan >>> origination) which extends the existing functionality with fineract as a >>> dependency. >>> >>> One thing is how will you link it. Client data is present in the core >>> fineract. How you will architect the plugin. and RBAC , flows, how will you >>> secure it? closer i can see it can be a separate module in loan >>> origination. in which we can access it. >>> >>> Another thing i want to confirm: is there any existing functionality for >>> this in core Fineract? "the Working Capital Loan documentation >>> <https://fineract.apache.org/docs/current/#_new_loan_product_type_working_capital_loan> >>> provides >>> a useful reference for the closest existing product." i think if there is >>> closest reference. If there is a close reference, it's better to keep it in >>> Fineract alone and not include it in GSOC, as this will give ample time for >>> FSIP or PMC discussion. >>> >>> Regards, >>> Aman >>> >>> On Mon, Nov 2, 2026 at 6:13 AM Abhishek Chaudhary < >>> [email protected]> wrote: >>> >>>> Hi all, >>>> >>>> As James suggested, I'm moving this discussion to the list. My original >>>> email about revolving loans (FINERACT-134) as a GSoC 2027 project and >>>> James's reply are below. >>>> >>>> Taking James's point that GSoC projects shouldn't keep adding >>>> functionality to core, I'd like to propose it as a plug-in instead: a proof >>>> of concept for a revolving line of credit built as a custom module ( >>>> https://fineract.apache.org/docs/current/#_custom_modules) in a >>>> separate public repository, with an FSIP drafted and discussed on its own. >>>> >>>> A few questions: >>>> 1. Would a revolving line of credit proof of concept as a custom >>>> module, outside core, be a suitable GSoC 2027 project? >>>> 2. The custom module docs describe modules as a proof of concept. Can >>>> they support a new product type today, or would part of the project be >>>> extending that mechanism? >>>> 3. Would anyone be interested in mentoring it? >>>> >>>> Thanks, >>>> Abhishek >>>> >>>> P.S. Thank you James for the detailed reply. I'd love to dig in and >>>> find out just how entertainingly complex the FSIP turns out to be. >>>> >>>> ---------- Forwarded message --------- >>>> From: James Dailey <[email protected]> >>>> Date: Mon, 2 Nov 2026 at 01:30 >>>> Subject: Re: GSoC 2027: Revolving Loans in Apache Fineract >>>> To: Abhishek Chaudhary <[email protected]> >>>> >>>> My perspective is that while this kind of functionality *may* be >>>> useful, I also think it should be part of a plug and play approach into >>>> Fineract. >>>> >>>> I also think our community should focus primarily on Security and >>>> Stability projects in Fineract code: paying down technical debt, adding >>>> tests, improving performance, and pruning functional areas. I think the >>>> *core* Fineract project should have less functionality out of the box, >>>> not more. That's my general and entirely personal take. GSOC projects >>>> that constantly "ADD" functionality into Fineract core have the side effect >>>> of making it more complex and potentially less maintainable. Proof of >>>> concepts (on outside public forks for a student or in separate branches or >>>> repos) are more welcome, in my view, than projects intended to go upstream >>>> into a system used in *Production* in a number of places. >>>> >>>> An FSIP for revolving line of credit is perhaps useful as a project in >>>> itself. So, I am certainly open to you proposing something for >>>> discussion. I'll be upfront and say that I may not be able to give much >>>> mentoring to this. Your short list is scratching the surface - the >>>> multitude of edge cases are "horrendously" or "endlessly" or >>>> "entertainingly" complex (depending on your point of view). >>>> >>>> - James >>>> >>>> >>>> >>>> On Fri, Oct 30, 2026 at 8:50 PM Abhishek Chaudhary < >>>> [email protected]> wrote: >>>> >>>>> >>>>> In the community chat, Adam mentioned that revolving loans would be a >>>>> useful addition to Fineract, with an FSIP once there is a high-level >>>>> design. >>>>> >>>>> I'd like to propose revolving loans as a potential GSoC 2027 project >>>>> and get your feedback before going deeper into the design. >>>>> >>>>> There is already some relevant groundwork: >>>>> - FINERACT-134 <https://issues.apache.org/jira/browse/FINERACT-134> asks >>>>> for revolving line of credit products, and James, your comment there >>>>> suggests an FSIP and outlines the main design concepts. >>>>> - FSIP-010 >>>>> <https://fineract.apache.org/docs/current/#_fsip_010_revenue_based_working_capital_loan_product> >>>>> shows >>>>> how the Working Capital Loan product was proposed as a new loan product >>>>> type. >>>>> - the Working Capital Loan documentation >>>>> <https://fineract.apache.org/docs/current/#_new_loan_product_type_working_capital_loan> >>>>> provides >>>>> a useful reference for the closest existing product. >>>>> >>>>> I have some initial thoughts on what the high-level design would need >>>>> to address: >>>>> - Credit limits and available credit >>>>> - Billing cycles, statements and minimum payments >>>>> - Interest calculation on revolving balances >>>>> - How the new product fits alongside the existing loan and working >>>>> capital loan modules >>>>> >>>>> A few questions I'd appreciate your thoughts on: >>>>> 1. Do you think revolving loans would be a suitable GSoC 2027 project? >>>>> 2. Would it make sense for me to work on the FSIP and high-level >>>>> design over the coming months, so that the GSoC project can focus on >>>>> implementing the first increment? >>>>> 3. What would be a sensible scope for that first increment? For >>>>> example, a basic credit line supporting drawdowns, repayments and >>>>> available >>>>> credit. >>>>> >>>>> >>>>> Getting the design right will require considerable research on my >>>>> part, so I wanted to get your thoughts on the direction and scope before >>>>> investing further in it. >>>>> >>>>> Thanks, >>>>> Abhishek >>>>> >>>>>
