Devntech

BACKGROUND

Please describe your company and position.

I am the CEO of Mambu

Describe what your company does in a single sentence.

We are the world’s only true SaaS cloud banking platform and the founders of composable banking.

Since 2011, we’ve helped banks, fintechs, lenders, credit unions and retailers design and launch modern financial products with speed, flexibility and scale.

Our composable approach works like building blocks: independent components, systems and connectors that can be assembled in any configuration to meet business goals and customer needs. From core banking and deposits to lending, payments and Islamic banking, we power it all.

Whether launching a single product or transforming an entire banking stack, we give organisations the agility, scalability and reliability to succeed in today’s financial landscape and tomorrow’s.

OPPORTUNITY / CHALLENGE

What specific goals or objectives did you hire DevnTech to accomplish?

• Build custom backend services and integrations that extend our core banking platform and connect reliably with third-party systems such as payments, KYC, and reporting tools used by our banking and fintech customers.
• Support faster customer product launches by offloading custom software work from our internal team while keeping full alignment with our platform rules, security standards, and banking logic.
• Deliver stable, production-ready code that can handle real transaction volumes, reduce issues during partner testing, and remain easy for our internal teams to maintain and update over time.

SOLUTION

How did you find DevnTech?

• Online Search
• Clutch Site

Why did you select DevnTech over others?

• Pricing fit our budget
• Good value for cost
• Company values aligned

How many teammates from DevnTech were assigned to this project?

2-5 Employees

Describe the scope of work in detail. Please include a summary of key deliverables.

DevnTech worked with us across several workstreams. The scope covered design support, backend development, API work, testing, and deployment support.

They helped us build custom services that connect our core banking engine with external systems used by our customers. This included payment providers, KYC tools, and reporting services. They also worked on product logic tied to loan workflows, account setup, and transaction handling.

Key deliverables included:

• Custom backend services built to our coding standards
• Secure APIs with clear documentation
• Data models aligned with our core platform
• Automated tests to support ongoing releases
• Support during sandbox testing with partner banks
• Fixes and refinements based on live usage feedback

They did not just hand over code. They stayed involved through testing and early use, which helped reduce issues once customers went live.

RESULTS & FEEDBACK

What were the measurable outcomes from the project that demonstrate progress or success?

The biggest result was speed. We were able to support more client launches at the same time without putting pressure on our internal team. Features that used to wait in the backlog were delivered and tested in active environments.

We also saw fewer issues during partner testing. The integrations built by DevnTech handled real transaction loads well and behaved as expected during edge cases, such as failed payments or partial data from third parties.

Internal teams reported less rework after handover. The code was readable and matched our platform rules, which made future updates easier.

Describe their project management. Did they deliver items on time? How did they respond to your needs?

DevnTech managed their work in a clear and calm way. Tasks were broken down well, and timelines were realistic. Most items were delivered when agreed, and delays were rare. When something needed more time, they explained why early and proposed options.

They were easy to reach and quick to respond. If our product team changed a rule or a partner raised a new need, DevnTech adjusted without friction. They joined planning calls, asked the right questions, and followed up with clear action points.

What was your primary form of communication with DevnTech?

• Virtual Meeting
• Email or Messaging App

What did you find most impressive or unique about this company?

What stood out most was their understanding of banking products. They did not treat this like a normal software job. They understood why rules matter, why data must be exact, and why small errors can cause big issues for banks.

They also blended well with our team. It felt like working with extended colleagues, not an outside vendor. They respected our standards and were careful with security and data handling.

Are there any areas for improvement or something DevnTech could have done differently?

At the start, onboarding took a bit of time due to the depth of our platform. This improved once shared docs and examples were in place. Beyond that, there were no major issues. Communication stayed strong, and quality stayed consistent.

BACKGROUND

Introduce your business and what you do there.

I'm the CEO of Domyn, an IT company that develops responsible AI solutions for regulated industries. We offer governable, composable AI architectures, including large language models (LLMs) and AI agents, supported by advanced supercomputing to help enterprises meet compliance and performance needs.

OPPORTUNITY / CHALLENGE

What specific goals or objectives did you hire Devntech to accomplish for your AI solutions?

We needed Devntech to build a secure AI workflow system with strong audit trails and access control. We also needed them to create a custom data pipeline that could handle large model inputs and rule checks. Additionally, we wanted them to develop a governance dashboard so risk and compliance teams could review AI outputs easily.

SOLUTION

What was the scope of their involvement?

Devntech has handled the full custom software build from early design to production release. They started with workshops with our AI, risk, and infrastructure teams. We mapped how our model runs, reviews, and approvals should work. After that, they built the system in modules so we could test each part early.

Key deliverables have included a custom AI job orchestration system for model runs, a secure data intake and validation layer, a full audit log for every model action and configuration change, role-based access with review and approval steps, a governance dashboard for compliance teams, an API layer for linking with our internal model tools, an admin panel for rule sets and control checks, and a test suite and staging setup for regulated clients.

They've also written clean system documentation so our internal engineers can maintain and extend the platform.

What is the team composition?

We work with a team of 6-10 from Devntech.

How did you come to work with Devntech?

We found them through an online search and the Clutch website. We chose them because their pricing fit our budget, they offered good value for the cost, and their company values aligned with ours.

What is the status of this engagement?

We worked together from January to May 2025.

RESULTS & FEEDBACK

What evidence can you share that demonstrates the impact of the engagement?

We track real numbers after going live. Our model review preparation time has dropped from about two hours to about 35 minutes per case. Our audit report preparation time has gone from a full-day task to under two hours. Our manual log checks have dropped from around 120 items per week to about 25. Our internal review tickets related to missing trace data have almost stopped.

One clear moment has stood out. During a client audit drill, we pulled a full model decision trail in minutes. Before this system, that would have required pulling logs from three tools and manual stitching.

As the client, we set clear KPIs at the start of the project, and Devntech has helped us hit or pass most of them. We track these from day one because our AI platform serves regulated sectors, so proof and trace matter.

Our main targets and achievements have been:

• AI decision trace coverage: Our target was to make every AI action traceable with full logs and review history. Before this project, only key steps were logged. After the new system, every AI job run stores full trace data. Each run shows the input source, rule checks, model version, and reviewer notes. Our audit lookup time has dropped from hours of manual digging to a few minutes using the search panel Devntech has built.
• Review and approval cycle time: We wanted faster reviews for flagged AI outputs. Our baseline was that a full review cycle often took 2–3 days with email and sheet-based tracking. After going live, most reviews now close within the same workday. Reviewers work inside one dashboard instead of four tools. Queue routing sends items to the right reviewer without manual sorting.
• Failed job recovery time: We set a target to cut how long it takes to detect and fix failed AI runs. Before, we found failures by checking logs manually. Fixing and rerunning often took half a day or more. After Devntech's build, auto alerts flag failed runs within minutes. One-click rerun works with saved configurations. Our mean recovery time has dropped to under one hour in normal cases.
• Admin workload per client workspace: We wanted each client AI workspace to need less manual setup and fewer support touches. The new workspace setup now takes about 30 minutes using templates. Support tickets tied to configuration errors have fallen sharply in the first two months. Most client teams now self-manage roles and rules using the admin panel.

We track these KPIs through system logs and admin dashboards and support desk data. Devntech has helped us define how to measure them, not just how to build the features. That has made the results easy to verify.

Another measurable result worth noting is how stable the system has stayed once it went live. After launch, we tracked system incidents, failed background jobs, and emergency fixes. In the first three months, critical incidents were very rare. No unplanned full outages occurred. Most issues were minor configuration mistakes, not code faults. Hotfix turnaround was often the same day when needed. That gave our operations and risk teams more trust in the platform.

We also measured how long it took new staff to get productive in the system. New reviewers were able to complete their first full workflow on day one. Our admin training time dropped to a single short session plus documentation. We no longer needed custom walkthrough calls for each new team.

We logged change requests and small feature additions. Many small requests shipped within one sprint. Larger feature additions had clear estimates and hit their target windows. We could plan releases with confidence because delivery stayed steady.

Before the project, we had limited visibility into AI job flow and control steps. After delivery, every job, rule check, and approval step became searchable. Audit preparation that once took days now takes a short working session. Internal audit reviews now use system exports instead of manual evidence packs.

We already covered the major KPIs earlier, but these delivery and quality metrics also show real progress. The value was not just in features shipped, but in how reliably and measurably the system improved over time.

How did Devntech perform from a project management standpoint?

Project management has been steady and simple. They work in short sprints with weekly demos. We always see real progress, not slides. When our compliance team asks for extra control points, Devntech adjusts the backlog and shows the impact on the timeline and scope.Most milestones land on the planned dates. When a task grows bigger than we first thought, they flag it early and give us clear options.

What did you find most impressive or unique about them?

They understand regulated AI work, not just code. Their team asks smart questions about audit trails, model risk, and review steps. They don't treat it like a normal app build. They treat it like control software.

Their engineers also write very readable code. Our in-house team reviews it and has no trouble following the logic.

Are there any areas they could improve?

We would have liked more sample UI mock-ups before the first build sprint. Once we asked, they added that step, and the flow improved. Aside from that, the work and teamwork have been strong.

How would you rate your overall satisfaction with the vendor?

From our client side, one last measurable point worth adding about working with Devntech is how clearly we can track progress during the build, not just after launch. We track planned versus delivered work each sprint. Most sprints close with all core tickets completed. Spillover tasks are usually low-priority items, not core features. Feature demos at the end of each sprint match the specifications we approve. That makes roadmap planning much easier for our product and risk teams.

We log defects found in QA and after release. QA defect counts drop with each test cycle. Repeat bugs are rare because fixes include added test cases. Post-release bug reports stay low and easy to trace using built-in logs.

Is there anything else the vendor did well?

We share real session recordings and notes. Devntech redesigns the rule builder with guided steps and presets. That closes the gap.

Is there anything that did not go according to plan?

From our client point of view, working with Devntech has been very positive overall, and most targets have been met. There are only a few areas where results come in a bit short at first, but they are handled well and improve in later updates.

One target is very fast load time for the audit and trace dashboard when large AI job logs are queried. In early releases, large queries take longer than we want. Power users who pull wide date ranges feel the delay.

Devntech reviews the query pattern, adds indexed search and cached views, and pushes an update. After that, load time improves, and the issue stops coming up in user feedback.

We hope non-technical risk staff can create complex rule chains on their own from day one. The first version works, but it feels too technical for some users.

System reports are correct but not client-ready in layout at the start. Teams still export and restyle them for board packs. That is below our expectation for “ready to share” output.

Devntech later adds report templates and layout options. That reduces manual rework a lot.

What stands out is their response. There's no pushback or blame. They reproduce the issue, explain the cause in plain terms, and ship fixes in planned sprints.

So yes, a few early targets are not fully met on the first pass, mostly around speed at scale and ease of use for non-technical users. None are left unresolved, and each one improves through follow-up releases.

BACKGROUND

Introduce your business and what you do there.

I’m the CEO of Bord Bia, The Irish Food Board. We connect Irish food, drink, and horticulture suppliers with customers worldwide, promoting exports and supporting industry growth through market insights and partnerships.

OPPORTUNITY / CHALLENGE

What challenge were you trying to address with Devntech?

We needed to build a stable e-commerce platform for multi-supplier product listings. We also wanted to make vendor onboarding simple with guided forms and checks, add buyer tools for bulk orders and quote requests, and connect the store with our product data and export records.

SOLUTION

What was the scope of their involvement?

Devntech built a full e-commerce system with supplier and buyer sides. They didn’t just design pages; they built the core order flow and vendor tools. The scope and key deliverables included a multi-vendor store with supplier dashboards, a product upload system with category and quality tags, a bulk order and quote request feature for trade buyers, an admin panel for approvals and supplier checks, order tracking with status notes and document upload, data sync with our product and export record database, and a new checkout flow with tax and shipping rules by region.Devntech handled more than just the technical build. They were involved from early planning through launch and post-launch support. 

What is the team composition?

We worked with 2–5 teammates from Devntech.

How did you come to work with Devntech?

We found them through an online search and the Clutch site. We chose them over other options because they had high ratings, their pricing fit our budget, they offered good value for the cost, and their company values aligned with ours.

How much have you invested with them?

The project cost is confidential.

What is the status of this engagement?

We worked together from June–November 2025.

RESULTS & FEEDBACK

What evidence can you share that demonstrates the impact of the engagement?

After launch, we tracked usage and order flow each month. Vendor signup time dropped from several days of email steps to one guided session. Order data errors fell from frequent weekly cases to rare edge cases. Buyer quote requests moved from inbox threads into tracked portal records. Our staff processed more supplier approvals per week with the same team size. Support tickets about product uploads dropped by over 40 cases per month. We verified this using admin logs, ticket tags, and order records.Before the new platform, vendor signup took about 2–3 business days on average. It involved email back-and-forth, manual document checks, and spreadsheet entry. After the new flow went live, most vendors completed signup in under six hours, and many finished the same day. The guided forms and document upload checks removed a lot of delays. Our team no longer had to chase missing files as often.With the old process, our staff handled about 25–30 supplier approvals per week. That number stayed low because each case needed manual review steps across different tools. With the new dashboard and approval queue Devntech built, we were able to process 60–75 supplier approvals per week using the same team size. The reason was simple: all records, files, and checks were in one place, and status flags showed what needed review next.Before launch, order records often had small but costly issues — missing supplier codes, wrong product tags, or incomplete shipping fields. We logged about 40–50 order data errors per month that required manual correction. After the new platform and validation rules were added, monthly order data errors dropped to around 10–15 cases. Most remaining issues came from supplier-side uploads, not system faults. Required-field checks and format validation blocked bad entries at submit time. Auto-mapping of supplier catalogs removed many SKU mismatch problems. We could see this directly in the admin error log Devntech built, which flagged rejected or corrected records.We also compared helpdesk tickets tied to orders and vendor onboarding. Before the update, we had 70–90 tickets per month related to order status confusion, failed submissions, or missing documents. After go-live, we had 25–35 tickets per month in the same categories. Most tickets shifted from “system not working” to simple usage questions. “Where is my order in the process?” tickets dropped a lot because users could now see live status in their portal.In the first full month after release, we ran a supplier onboarding drive with over 120 new vendors. Under the old setup, that level of activity would normally overload support. With the new system, ticket volume stayed stable, and approvals moved faster, which confirmed the fix worked under real load. So yes, both error counts and ticket logs gave us solid, trackable proof that the new build improved daily operations.

How did Devntech perform from a project management standpoint?

Devntech’s project manager ran weekly calls with clear task lists. We always knew what was in progress and what was next. When we changed rules for export documents mid-build, they adjusted the plan and showed the impact on time and cost right away. Most items came in on the planned date. When a task slipped, they explained why and showed the fix plan the same day. Messages were short and clear.

What did you find most impressive or unique about them?

Devntech understood that our users were not tech experts. Many suppliers were farmers or small food brands. Devntech kept forms simple and added help text in plain words. They tested flows with sample supplier data, not fake demo content. That made a big difference.They also built strong admin filters. Our team could now find products, suppliers, and orders in seconds.

Are there any areas they could improve?

Early design drafts had too many admin options on one screen. We asked for simpler layouts. They fixed it, but it took an extra design round. A shorter first draft would have saved time.

BACKGROUND

Introduce your business and what you do there.

I’m the co-founder and CEO of Comodule, a company that provides connectivity solutions for e-bike and e-scooter brands. We enable real-time internet access for light electric vehicles to support emission-free urban transportation.

OPPORTUNITY / CHALLENGE

What challenge were you trying to address with Devntech?

We needed Devntech to build a new web control panel for our vehicle brand partners with faster load times and a clearer device data view. We also needed them to create custom backend services to handle high device message flow without queue jams. Additionally, we wanted them to rework parts of our partner portal so our support teams could manage device fleets with fewer manual steps.

SOLUTION

What was the scope of their involvement?

Devntech handled both custom software work and web development. Their scope covered backend services, partner dashboard features, and system links with our device data pipeline. Key deliverables included a rebuilt partner web portal with new device list views, search tools, and live status panels; new backend services for device message intake, job routing, and error logging; admin tools for our support team to reset devices, push config updates, and review trip logs; API links between our device cloud and the partner dashboard; and a job monitor screen that shows stuck or delayed device sync tasks in real time.One useful feature they built was a device replay tool. When a brand partner reported a bad trip record, we could replay the message stream and see where the break happened. That saved a lot of back-and-forth email.Devntech focused most on the backend services and partner web portal since those were the areas causing the most strain in our system. They spent most of their time on building and stabilizing the new backend message intake and job routing services, reworking the partner dashboard and admin portal, creating device management tools for our support staff, improving API links between our device cloud and the web interface, and adding logging, error tracking, and replay tools for device data issues.They didn’t design or build the IoT hardware or firmware on the vehicles. They also didn’t manage our core device connectivity network layer, handle UI brand design or marketing website content, or run long-term hosting or DevOps after the final rollout. Their role stayed focused on custom software and web platform engineering, which matched what we hired them to do.

What is the team composition?

We worked with four teammates from Devntech.

How did you come to work with Devntech?

We found them through an online search and the Clutch site. We chose them over other options because they had high ratings, a great culture fit, and company values aligned with ours.

How much have you invested with them?

The project cost is confidential.

What is the status of this engagement?

We worked together from March–September 2025.

RESULTS & FEEDBACK

What evidence can you share that demonstrates the impact of the engagement?

We tracked results across support, system load, and partner use. The average dashboard load time dropped from around nine seconds to under three seconds for large fleets. Device message queues that once held over 120,000 pending records during peak hours now stayed under 15,000. Support tickets tied to “device not updating” cases dropped from about 40 per week to under 20.Partner logins increased, with over 300 daily active partner users after the new portal launch. Our support team saved several hours each day using the new admin tools instead of manual database checks.We measured the increase in partner logins using two simple metrics from our portal analytics and auth logs. First, we tracked unique partner user logins per week. This counted how many distinct partner accounts signed in to the dashboard at least once during a seven-day window. After the Devntech rebuild, that number rose steadily because more partner teams started using the tools instead of asking support for reports.Second, we tracked total login sessions per day. This showed how often users came back, not just how many showed up once. We saw more repeat sessions from the same partner companies. Many support agents and fleet managers began logging in multiple times during their workday to check vehicle status and connection logs.We also cross-checked this with portal audit logs, session records from our auth service, and support ticket tags that showed whether a case was solved through the portal or by email.We set a few clear KPIs before the project started and compared results to our past monthly averages to judge success. Before Devntech began, we defined simple, trackable targets tied to daily use and support load. The main KPIs were a higher number of weekly active partner users in the portal, more self-serve actions completed without staff help, faster task time for connection checks like vehicle status and device sync, and a lower volume of repeat support tickets on login and data access issues.We pulled three months of baseline data from our logs and support desk reports. That gave us a normal monthly average for logins, sessions, and ticket topics. After launch, we reviewed the same data points each month and compared them side by side with that baseline.Some gains were planned against targets, like partner portal usage and task time. Others were judged against the past averages, like ticket drops and repeat session counts. Using both methods made the results easier to trust and explain to our internal team.We used a mix of simple dashboards and system logs to track each KPI and compare results with our baseline data before the project with Devntech started for Comodule. We didn’t rely on one tool. We combined product analytics, server logs, and support data so the numbers matched across sources.We used a product analytics dashboard to track partner portal activity, unique users, session counts, and feature usage. We built custom panels for weekly active users and repeat visits. We used auth and backend logs to confirm login events, session length, and failed vs. successful access attempts. This helped verify the analytics data.We also used a support desk dashboard to track ticket volume by tag and topic. We compared login and device data tickets before and after launch. Our data team pulled monthly summary tables so we could compare new numbers against the three-month baseline period.We reviewed these dashboards every two weeks during rollout, then monthly after full release. Because the same KPIs showed the same trend across tools, we felt confident the gains were real and well tracked.

How did Devntech perform from a project management standpoint?

Their project flow was clear and steady. We had weekly check-ins and short mid-week status notes. Tasks were tracked in a shared board. Due dates were visible and mostly met.When we changed a feature mid-build, they didn’t push back. They reviewed the impact, gave a new time note, and adjusted the sprint plan. Urgent bugs were often fixed within one or two days.

What did you find most impressive or unique about them?

The most impressive part was how fast they understood our device data flow. IoT vehicle data is messy. Messages arrive late, out of order, or in bursts. Devntech designed guards and retry rules that handled this without complex setup on our side.They also tested with sample device streams we gave them, not just mock data. That caught edge cases early.

Are there any areas they could improve?

Early in the project, their tech notes were a bit too brief for our internal audit team. We asked for longer change logs and API notes. After that request, they added better docs and short video walkthroughs. That solved the gap.

BACKGROUND

Introduce your business and what you do there.

I’m the co-CEO and CIO of Heimstaden AB, a residential real estate owner and industrial partner. We leverage extensive strategic and operational expertise in residential real estate and ESG to create enduring value.

OPPORTUNITY / CHALLENGE

What specific goals or objectives did you hire Devntech to accomplish for Heimstaden?

We needed Devntech to build custom internal software to support property, tenant, and operational data in one place. We also wanted to improve reporting and visibility so teams in different regions could access the same reliable data. Additionally, we wanted to reduce manual work tied to asset tracking, updates, and internal coordination. Lastly, we wanted to support long-term scale as our portfolio continues to grow across markets.

SOLUTION

What was the scope of their involvement?

Devntech worked closely with our internal teams to understand how our operations run across countries and business units. They helped design and build custom software that fits our processes instead of forcing us to change how we work. Devntech’s scope covered backend development, system logic, integrations with existing tools, and support during rollout. They focused on building systems that are stable, clear, and easy for our teams to use day to day.

What is the team composition?

We worked with 3–4 teammates from Devntech.

How did you come to work with Devntech?

We found them through an online search and the Clutch site. We chose them because their pricing fit our budget, they were a great culture fit, and they offered good value for the cost.

How much have you invested with them?

The project cost is confidential.

What is the status of this engagement?

We worked together from May 2025–November 2025.

RESULTS & FEEDBACK

What evidence can you share that demonstrates the impact of the engagement?

Teams across regions were able to access consistent data without relying on spreadsheets or email updates. Reports that once took days to prepare could be reviewed quickly using shared dashboards. Operations teams spent less time fixing data issues and more time focusing on asset quality and tenant needs. The system handled daily updates smoothly, even as data volume increased. From a management view, it became easier to get a clear picture of assets without chasing information across teams.

How did Devntech perform from a project management standpoint?

Devntech ran the project with structure and discipline. Timelines were realistic, and deliverables were tracked clearly. Most work was delivered on schedule, and follow-up fixes were handled without delay. When requirements changed, which is common in large property groups, they adjusted calmly and explained options clearly. Communication was direct and practical.

What did you find most impressive about them?

Devntech took time to understand our business. They didn’t rush into building features without knowing how they would be used in daily operations. They also showed strong judgment when dealing with data accuracy and access control. That level of care was critical in a company managing a large residential portfolio.

Are there any areas they could improve?

No, there weren’t any.