About Atmora
Engineers who have run
the thing they built
Roughly 120 engineers, designers and delivery leads working in small, senior, long-lived teams — most of whom have spent part of their career on the operator's side of the desk.
- 10
- years in production
- 120+
- engineers and designers
- 14
- countries shipped into
- 240+
- projects delivered
We are not a body shop, and we are not a design studio that outsources the build. Atmora Tech takes on systems where correctness matters and downtime is measured in lost revenue rather than lost engagement: auction and bidding engines, quote-to-cash, trading and risk, warehouse and yard operations, procurement.
That shapes what we say yes to. We work in two-week increments against a written definition of done, we deploy to production in the first three weeks of every engagement, and we hand over documentation your own team can actually maintain. If we cannot commit to that, we will say so in the first call rather than the fourth invoice.
Principles
How we decide
Production deployment in week three
Not a staging environment and not a demo. By the end of week three a real slice of the system is deployed to production in your cloud account, behind a feature flag, with Prometheus alerts and an on-call runbook. Pipelines, secrets management and rollback are proven with live traffic before anyone writes the second feature. If we miss that date, it appears on the week four steering report with the reason.
Named engineers, held by contract
The statement of work lists the team by name, seniority and allocation percentage. Any change requires 30 days' written notice and a two-week overlap handover, and the replacement is proposed to you rather than assigned. Over the past three years, 78% of engineers stayed on an account for its full duration; on our four longest-running clients, the original tech lead is still in place.
Discovery you can walk away from
Two to three weeks, fixed fee, no obligation to continue. You receive the architecture assessment, the risk register, the migration sequence and a costed estimate with an explicit ±25% band, all under your copyright. Roughly one discovery in six ends with us recommending you do less than you asked for, or nothing at all. We have written that recommendation for clients who were ready to sign eight-figure programmes.
Your repositories, your cloud, from the first commit
Code lands in your GitHub organisation and infrastructure in your AWS, Azure or GCP accounts on day one. There is no Atmora-hosted runtime, no proprietary framework you have to keep licensing, and no source code escrow arrangement, because there is nothing to escrow. If you terminated tomorrow, the system keeps running and your team keeps deploying it.
Delivery metrics published every month
Every steering pack carries the DORA four per team — deployment frequency, lead time for change, change failure rate, mean time to restore — pulled from your CI and incident tooling rather than from a slide. Current portfolio medians: 11 deployments per week, 2.4 days lead time, 6% change failure rate, 41 minutes to restore. You can audit the source data yourself; it sits in your systems.
Modernisation without a code freeze
We replace legacy systems with strangler-fig routing: old and new run side by side, traffic moves capability by capability, and each cutover is reversible by a routing flag in under ten minutes. Across more than 40 modernisation programmes since 2016, none has required a business change freeze longer than 48 hours, and no cutover has been rolled back more than once.
Timeline
Ten years, abbreviated
- 2016
Four people and one contract
Atmora started in a first-floor office in Ratanada, Jodhpur, with four engineers and a single contract: a stock-reconciliation tool for a regional distributor running 14 depots on spreadsheets. It shipped in eleven weeks and is still in use, which is either a compliment or a warning.
- 2018
Manufacturing work and the Bengaluru office
Kestrel Industrial asked for a manufacturing execution system across two plants. Delivering it meant learning OPC-UA, shift-pattern data and the reality that a plant floor does not stop for a deployment window. Headcount went from 20 to 38, and we opened in Bengaluru to hire the platform and data skills Jodhpur could not yet supply.
- 2020
Remote delivery, tested properly
Brightmoor Foods lost its distributor channel in March and needed direct order capture. We rebuilt their order and allocation flow in eleven weeks with a team that never met in person, peaking at 4,100 orders a day against a previous ceiling of about 600. The working practices from that project — written decisions, asynchronous review, recorded demos — became the default.
- 2021
London and the first regulated client
We registered a UK entity and opened in London to be in the same room as clients who cannot outsource their architecture decisions. Cordage Bank was the first, starting with a six-week assessment of a core banking periphery built on a shared Oracle schema that 31 applications wrote to directly.
- 2023
Cloud and platform practice formalised
Ad-hoc infrastructure work became a practice with its own standards: a shared Terraform module library, Argo CD for every deployment, and a policy that no engagement goes live without dashboards and a runbook. Corvus Logistics was the proving ground, with a two-region active/passive setup and a tested 22-minute failover.
- 2024
AI work with an evaluation bar
We took on applied AI on one condition: no model ships without a labelled evaluation set the client has signed off. The first two engagements were document extraction for Orbit Supply, at roughly 30,000 pages a month, and price-anomaly detection over Northwind Commodities' trade blotter. Both replaced manual review rather than adding a layer on top of it.
- 2026
120 people, three offices, ten years
Around 120 engineers, designers and delivery leads across Jodhpur, Bengaluru and London, working with clients in manufacturing, logistics, financial services, commodities and procurement. ISO 27001 certified, SOC 2 Type II reported, and still small enough that the people who wrote the code answer the phone when it breaks.
Leadership
Who runs it
Aditya Rathore
Founder & Chief Executive
Started Atmora in 2016 after nine years building order-management and distribution systems, mostly in Delphi and PL/SQL, which he will still defend given the chance. He handles the commercial conversations and, more often, the ones where a project has gone wrong. Reads every post-incident review the company produces.
Neha Vyas
Chief Technology Officer
Joined in 2019 from six years on payments infrastructure, where she learned what a reconciliation failure costs when it is discovered a week late. She sets the technical standard across the three offices and chairs the architecture review that any engagement over £250k has to pass. Wrote the first version of the deployment pipeline and has been trying to delete it ever since.
Malcolm Ashcroft
Managing Director, UK & Europe
Eighteen years in financial services technology, including four running application delivery inside a UK clearing bank, which is where he formed his views on change control. Opened the London office in 2021 and now leads the regulated-sector business. He is the one who tells clients that their migration will take longer than they have been promised elsewhere.
Priya Raghavan
Director of Client Delivery
Runs delivery across all three offices, roughly 30 concurrent engagements at any point, and owns the staffing model that keeps them fed. She came up through business analysis and project management on ERP rollouts, so she reads an estimate with appropriate suspicion. Her weekly portfolio report names slipping projects in the week they slip, not the month after.
Karan Malhotra
VP, Cloud & Platform Engineering
Built and now owns the internal platform: the Terraform module library, the Argo CD layout, and the paved path that gets a service from merge to production in under 12 minutes. Previously ran infrastructure for a logistics SaaS through a migration off self-managed Kubernetes and back onto it two years later, an experience he describes as instructive. Argues frequently that a pair of VMs would do.
Ishita Bose
Head of AI & Applied Research
PhD in statistical machine learning, then four years applying it to industrial time-series data where the labels were scarce and the sensors were unreliable. She set the rule that no model leaves Atmora without an evaluation set the client has agreed to, and she enforces it. Spends more of her time on measurement and failure analysis than on modelling, which she considers the correct ratio.
Offices
Where we are
Jodhpur
India
Ratanada, Jodhpur, Rajasthan 342011
Bengaluru
India
Indiranagar, Bengaluru, Karnataka 560038
London
United Kingdom
Farringdon, London EC1M
Careers
8 open roles
Small teams, senior people, long engagements. We hire slowly and keep people for years.
Bengaluru, India · Full-time — hybrid, three days a week on site
Senior Backend Engineer (Java)
You will work on the settlement and ledger services behind Meridian Exchange's B2B marketplace: roughly 40 million ledger rows a month, Java 21 on Spring Boot, Postgres 16, Kafka for event fan-out. The interesting part is not new endpoints, it is correctness under load — idempotent writes, exactly-once effects on top of at-least-once delivery, and holding p99 write latency under 250 ms as the ledger grows.
Jodhpur, India · Full-time — on site
Platform Engineer, Kubernetes
Atmora runs more than 40 client workloads across EKS and AKS, from a nine-node cluster carrying Kestrel Industrial's shop-floor telemetry to a two-region active/passive setup for Corvus Logistics. You will own the internal platform those delivery teams deploy onto: the Terraform module library, the Argo CD app-of-apps layout, and the paved path that takes a service from merge to production in under 12 minutes.
London, UK · Full-time — London office, with four to six client-site days a month across the UK and EU
Principal Solutions Architect
Most of our UK work is legacy modernisation for regulated businesses: Cordage Bank moving a 22-year-old core banking periphery off a shared Oracle schema, Aldridge Risk replacing an overnight batch that had grown to 9 hours and 40 minutes against a 6-hour window. You will run discovery, decide the migration shape, and stay accountable for it once delivery starts — not hand over a slide deck and leave.