EmpCloud — Delivery intelligence

Ranks measured GitHub + Hubstaff signal only — not a performance review, not developer worth.

Generated by an AI model reading commit/PR content, complexity and tracked time. It cannot see DevOps, incident response, mentoring, architecture, code review depth, or work outside the connected repos — read each person's caveats before drawing any conclusion about them.

Conclusion

claude-opus-5 · period Thu Aug 27 2026 00:00:00 GMT+0000 (Coordinated Universal Time) – Sat Sep 26 2026 00:00:00 GMT+0000 (Coordinated Universal Time) · generated 2026-09-26 08:50 UTC · triggered by cron:ec2 · 22/24 people had evidence to rank on · ~$0.834

**The period was heavy on shipping and light on process.** Nobody opened or reviewed a PR, so there is no review trail anywhere in this data. That makes commit titles and AWS events the only real evidence, and it caps how confident any of this can be.

**Harsh Patel and Pankaj Kumar stood out** because both surfaces line up: substantive application commits plus sustained, real infrastructure changes across 18 distinct days. Aziz Bohra carried the email-service build with the clearest fix and auth titles of the group. Rushendra Varma ranks high on infrastructure alone, with 20 active days and the widest service spread, which the policy explicitly supports.

**Watch the volume traps.** Several people sit far above the 35-commit median on repeated site scaffolds, repo dumps or "initial commit" messages, and several high infra counts are the same call repeated during deploys. Those are noted in each person's caveats. A few people tracked full months with no visible artifact at all, which usually means unmapped identities rather than no work, but it needs checking.

Ranked (GitHub + Hubstaff evidence, evaluated by Claude)

Rank reflects measured delivery signal in this dataset, not overall developer worth — read the caveats.

#	Name	Tier	Confidence	Rationale	Caveats
1	Harsh Patel	top	0.40	Strong on both surfaces. On code, 102 commits include real product work like "i18n (EN + PT-BR) Frontend", "Trend ingestion Phases 2, 4, Apify sources, content briefs, scale tuning" and "Add Nexus Service external integration API with live progress polling", plus performance work in "code splitting, caching, and favicon fixes across all apps". On infrastructure, 1085 write events across 18 distinct days and 19 services, with Elastic Beanstalk environment updates, auto scaling group changes and load balancer target registration, which is sustained deploy and scaling ownership rather than a one day burst. NexusAI@0bd983e NexusAI@0432529 NexusAI@009760f coxa-1touch@ec0b329 NexusAI@7b527dd	1085 infra events is by far the highest in the team, but much of it is the repeated suspend, update, resume pattern of a deploy cycle, so read it as deploy cadence and not as 1085 separate decisions. Some commit titles are vague about what actually changed, for example "pod setup and gpu changes draft" and "facefusion + heygen fixes". No PRs opened or reviewed, so nothing here was peer checked in GitHub. Infra work is concentrated in the 1touch Cloud Infrastructure account. Confidence is 40 percent because no task tracker confirms this was the requested work.
2	Pankaj Kumar	top	0.55	The clearest infrastructure record in the team, read by content and not count: Amplify app, branch and job creation, IAM role and policy creation with AttachRolePolicy, and Route53 hosted zone creation, spread over 18 distinct days and three accounts. The code side backs it up with real integration work rather than scaffolding, including "Add isolated TheSportsDB proxy with Mongo cache and Postman collection." and "Implement RemoteVozApi for editorial content and enhance sports data". 71 commits, 35 of them refactor, sits just above the 35 commit median with a healthy mix. cms@26be58b Brazil_CMS@a20ec6c lavozec.com@5d62e02 ecuadorenvivo.com@4bf8c96 puntoecuador.world@1bce1ac	1519 infra events is the highest number here, and a large share is repeated Amplify StartJob and Route53 record changes during site launches, so the event count overstates the number of distinct decisions. Several commits are new site setups that look similar across repos, for example "Initial commit from Create Next App" in puntoecuador.world. IAM role and policy creation is security sensitive and has no review trail in this data. No PRs opened or reviewed this period.
3	Aziz Bohra	top	0.40	Carried a large share of the email-service build with titles that say what changed: "Add session-cookie login/logout alongside Bearer API-key auth", "integrate auth readiness controls" and "prove mailbox recovery reconciliation". 180 commits is five times the median, but the mix is not padding, with 21 test commits and real refactor work in "Polish frontend UI migration and fix integration harness". Infrastructure matches the product, with 161 write events over 12 distinct days doing Route53 record changes and SES identity and DKIM setup, which is the domain onboarding this service needs. email-service@8cc9090 email-service@63ffd78 email-service@9b23c87 HyerPathEnrichment@1f52b81 email-service@8270b19	Several titles are labelled by day rather than by change, for example "align Day 8-10 manager-only workflow", which hides the actual scope. 180 commits at five times the median needs a sanity check for splitting, although the sample titles read as distinct pieces of work. All infra activity sits in the 1touch Cloud Infrastructure account and is mostly the same DNS and SES pattern repeated per domain. No PRs opened or reviewed, so none of this was peer checked.
4	Rushendra Varma Mudunuru	top	0.40	Zero commits, but the infrastructure record is the most sustained in the team: 285 write events across 20 distinct days and 17 services, which is the widest service spread here. The events read as account and platform ownership rather than deploy noise, including CreateAccessKey and TagUser in IAM, security group rule changes, service quota increase requests and AWS support case communication. Under the current policy that is a top contribution on its own.	No commits or PRs at all, so there is nothing to cross-check the infra work against. A repeated block of ModifySecurityGroupRules on the same security group inflates the event count more than it shows distinct work. Access key creation and user tagging are security sensitive actions with no review trail in this data. All 153.4 hours landed on one Hubstaff project that is not mapped to a repo, so time attribution tells us nothing about what was built.
5	Anshuman Parmar	top	0.55	Built the delivery reporting platform itself, with "Delivery intelligence platform: ingestion, scoring, and evidence UI" and "Add Notion as a work source, plus a Tasks page and per-person flags", plus data foundation work in "Complete Phase 0 data registry, schemas, and S3 load/validate/versioning". On infrastructure there are 302 write events over 11 distinct days across 12 services, including Elastic Beanstalk application versions and environment updates and clean-up work such as DeleteStack, DeleteSecurityGroup and DeleteLaunchTemplate, which is real teardown, not just deploys. Hubstaff-GitHub-Monitoring@61e805f Hubstaff-GitHub-Monitoring@d575cd6 POSITION_INTELLIGENCE@d33cc15 Hubstaff-GitHub-Monitoring@13a6998 Social_Media_Automations@de000f2	"Added Dashboard" carries 45052 churn on a title that says almost nothing about what was added. The same title "Align theme to POSITION Brand Guidelines v1.0; fix all-grey regression" appears twice in the sample, which suggests duplicated or re-applied work. 280 of 302 infra events sit in "account 1" rather than the main infrastructure account, worth confirming that is intended. No PRs opened or reviewed this period.
6	Abhishek Kulkarni	solid	0.55	140 commits, four times the median, and the largest ones are genuine foundation work: "establish PulseDesk engineering foundation (NOR-31)" at 8882 churn, an architecture and implementation plan covering DB schema and API design, and "add provider-neutral deployment manager". There is real fix work too in "close finished mixed Ask jobs and stop stale Need...". Infrastructure is modest but real, with a backup bucket created and EC2 key pair and instance provisioning across 7 active days. PulseDesk-Test@35f4be9 1touch-wave2-test@2009b9b 1touch-wave2-test@9dc9f1a ZeroHumanMultiAgentV3@2a540a3 ZeroHumanMultiAgentV3@e8f2c50	A good part of the 63 infra events are ConsoleLogin and CheckMfa sign-in records rather than changes, so the real infra footprint is smaller than the number suggests. The commit sample is dominated by a handful of very large commits, which makes it hard to tell what the remaining 130 commits contained. Two commits describe the same architecture plan, one as a plan and one as the implementation, so scope may overlap. No PRs opened or reviewed this period.
7	Ankur Kumar	solid	0.55	108 commits with a decent spread of real features: "migrate Nexus storage to AWS S3", "render episodes with 16:9 H3 motion and video lipsync" and an internship platform with apply flow and university partnership. Infrastructure shows 96 write events over 13 distinct days, which is good consistency, though the sample is almost entirely security group rule changes on two groups. Solid delivery, held below the top by repetition on both surfaces. NexusAI@db85c53 NexusAI@48a38ec coxa_crm@4d22e8b NexusAI@8b40863 NexusAI@4b0162b	Two commits in the sample share the exact same title and churn for the POSITION campaign integration, which looks like a duplicated commit rather than two pieces of work. The infra sample is dominated by repeated ModifySecurityGroupRules on the same two security groups, which is low variety for 96 events. 252.5 hours is the highest tracked time in the team, spread over five projects. Hours are not credit and this does not lift the rank. No PRs opened or reviewed this period.
8	Dhiraj Ambekar	solid	0.55	39 commits, close to the 35 median, but the titles are unusually clear about scope: "Add gated POSITION demo with passwordless email OTP" at 10721 churn, "Add admin projects, S3 video uploads, and video-play audit" and "bulk edit UI for personas, bots, and campaigns". Infrastructure is small but varied and spread out, with 49 write events over 9 distinct days across 11 services, including an Elastic Beanstalk environment update and Route53 record changes. Good quality per commit rather than volume. sma-presentation-gate@2b8b5b7 sma-presentation-gate@9b94baa Social_Media_Automations@8392093 Social_Media_Automations@4e5c0bd	"update the current code of dummy number" is a placeholder-style title on a 1482 churn change. Volume is at the median, so the case rests on the quality of the titles rather than breadth of delivery. Infra activity is split across two accounts, with most events in "account 1". No PRs opened or reviewed this period.
9	Jitu Prosad Saha	solid	0.40	Exactly at the 35 commit median, but the email-service work is well scoped and backend-heavy: "add SES bounce complaint suppression handling", "add outbound relay and per-tenant send quotas", "add Stalwart password and SES DKIM integration" and "add inbound pipeline reconciliation and evidence harness". This reads as someone owning deliverability plumbing, which is hard to get right. Small matching infra footprint with Route53 and SES identity setup. email-service@68cfba8 email-service@0aabc8d email-service@2ee5433 email-service@9dcaa85 email-service@23c8f0c	"add production backups and monitoring skeleton" appears twice with nearly identical churn, which looks like a duplicated commit. Only 13 infra events on 2 active days, and some of those are setting up his own MFA device rather than shared infrastructure. Volume is at the median with no PRs, so there is limited evidence of breadth beyond the email service. No reviews given or received in this data.
10	Rushikesh Powar	solid	0.40	55 commits above the median, with a mix of integration and clean-up: "Integrate panaromaec article API with canonical endpoint mapping", "Remove legacy PostgreSQL collection runtime" and "Ship network SEO remediation for all 25 public sites (phases 0-5)." Removing a legacy runtime is the kind of work that rarely shows up as a feature but matters. Infrastructure is light but sensible, with a new S3 bucket plus PutBucketEncryption and Route53 changes over 8 active days. panoramaec.com-@75c1116 POSITION_INTELLIGENCE@6a4ca9c Global_CMS@0409205 EcuadorCentral.com@a3880bb	"update" on an 8055 churn change is a placeholder message and hides what was actually done. Repeated StartInstances on the same EC2 instance across several days is routine operation, not infrastructure change. Two commits describe integrating six websites each and look like the same job repeated per batch. No PRs opened or reviewed this period.
11	Akash Kumar	solid	0.55	The real work here is on email-service and it is decent: "Admin portal real screens (login, tenants, domains, mailboxes)", "billing usage dashboard and gated ops screens" and a proper "regression test for stale-host-validator image-key rejection". Infrastructure is small but purposeful, with SES identity creation, DKIM attributes and Route53 record changes on 4 active days, which lines up with the email product. Held here because the headline churn is misleading. email-service@2884004 email-service@445a2c8 email-service@74c0116 server-codebase-backup@0957d77 server-codebase-backup@df0b34d	The two largest commits by far are server codebase backups at roughly 595k and 355k churn. That is a repo dump, not authored work, and it should not be read as output. Several titles are labelled "Day 3", "Day 4" and "day 6", which describe when rather than what. No PRs opened or reviewed this period. Hours are split across three projects, none mapped to a repository.
12	Manish Chandrakar	solid	0.40	126 commits, well above the median, spread across many language CMS sites. The best of it is real: "Add Italy sports hub, harden Serie A APIs, and soft-fail flaky upstream", "Switch sports data to TheSportsDB and add Japan top sports." and "Add production SEO, GEO, and pre-commit lint/build checks." But a lot of the volume is the same template rebranded per market, so the count runs ahead of the distinct problems solved. global-cms-Italiano@dbe84da global-cms-Japanese@ce3d982 ecuadorenvivo.com@1b70e1a EcuadorCentral.com@a1f9bbb global-cms-Korean@4224366	"first commit" at 8036 churn and "husky test" at 8018 churn are placeholder messages on very large changes, which is the pattern the ranking is meant to catch. High commit volume with repeated rebrand and palette work across sites should not be read as five markets worth of new engineering. Zero infra events, which may simply mean no AWS identity is mapped for him, not that no infra work happened. No PRs opened or reviewed this period.
13	MOHD NAVED	solid	0.40	Only 19 commits, well under the median, but they are among the best described in the team and lean toward hard correctness problems: "queue offline writes, add Club/FanBox/POS PWAs, and fall back", "survive API timeouts and brief offline without fake writes" and "resolve QA fan portal bugs BF-1, BF-2, and BF-3". There is also privacy work in "DSR cross-system deletion, redaction, and cloud baseline". Low volume, high care. coxa-1touch@cf96cab coxa-1touch@ff64930 coxa-1touch@ccfaa71 coxa-1touch@c8b12b4 POSITION_INTELLIGENCE@ed2fecf	19 commits against 201.5 tracked hours is a wide gap, so a lot of what he did this period is probably not visible here. His only 3 infra events are setting up his own MFA device and password, which is account hygiene and not infrastructure work. No PRs opened or reviewed, so the QA bug fixes have no review trail. All hours sit on projects not mapped to a repository.
14	Rishav Raj	solid	0.40	Only 6 commits, far below the median, but every one is a feature and they are large and coherent: "complete migration phases" at 14577 churn, "add content aggregation intelligence workflow", "add Google Search scraper support" and "add complete Custom Scrapers frontend workspace". This reads as a few big, well-scoped pieces rather than as doing little. Activity percentage of 42.3 is the second highest in the team. POSITION_INTELLIGENCE@a78df0d POSITION_INTELLIGENCE@95015d4 POSITION_INTELLIGENCE@09edede Social_Media_Automations@6879d25	"complete migration phases" carries 14577 churn on a title that does not say what was migrated or to where. 6 commits against 199.1 tracked hours is the largest gap in the team, so most of the period is not represented in this data. Zero infra events, which may mean no mapped AWS identity rather than no infra work. No PRs opened or reviewed, so none of these large changes were reviewed in GitHub.
15	Pramod Soni	solid	0.40	30 commits, just under the median, all following one consistent pattern of adding sports sections to language CMS sites: Vietnamese football routes, a Punjabi cricket section with fixtures and teams, a "Russian football hub via TheSportsDB CMS proxy." and a Tamil cricket section. The work is clearly described and repeatable, but it is the same job applied across sites rather than new ground, and the churn per change is modest. global-cms-veitanamese-new@a993ee5 global-cms-punjabi@ecb41f1 global-cms-Russian@34676db global-cms-Russian@b6d9445	Almost every commit is the same feature template repeated for a different language site, so breadth of problem solving is limited. "Update metadata and content for improved SEO and user experience" is broad and hard to check. Zero infra events, which may mean no AWS identity is mapped for him rather than no infra work. No PRs opened or reviewed this period.
16	Pranav Ojha	needs-attention	0.40	35 commits at the median, but the sample is dominated by four separate commits titled just "initial commit", each between 11000 and 12600 churn, which are site scaffolds rather than considered changes. There is genuine work underneath, notably "Dev Console created for intelligence ops" at 9596 churn and SEO and layout corrections across the Hindi and Indonesian sites. 32 of 35 commits fall into the "other" category, so the record is hard to read. lavozec.com@2c1c466 informeec.com@69d8ff7 POSITION_INTELLIGENCE@0ee7869 global-cms-hindi@fba9fb9	Four "initial commit" messages with very large churn make it impossible to tell what was written versus generated from a template. Nearly all commits are uncategorised, with no fixes, tests or refactors identified. His 3 infra events are his own password change and MFA setup, not infrastructure work. No PRs opened or reviewed this period.
17	Cursor Agent	needs-attention	0.07	An automation identity, not a person, with 6 small commits on email-service that form one coherent chain around domain verification: "Add manual domain DNS verification", "Keep manual verified domains on public DNS checks" and "Promote domain after representative mailbox create". Churn is tiny, between 35 and 515, so this is fine grained follow-up work attached to someone else's feature. email-service@b9bc59d email-service@441362f email-service@048305b	This is a tool account, so the credit for this work belongs to whoever drove it, and that person is not identifiable from this data. No tracked hours, no infra events and no PRs. Evidence confidence is 7 percent, the lowest available.
18	Claude Code	needs-attention	0.07	An automation identity with a single 44 churn commit, "Fix AttributeError crashing entire Gemini structuring batches". The title is specific and describes a real crash fix, but one small commit is the entire record for the period. Football_Scraping_Main@5f168e4	This is a tool account, so the human who directed the fix is not visible here. No tracked hours, no infra events and no PRs. Evidence confidence is 7 percent, the lowest available.
19	mohammad saad	needs-attention	0.40	173.5 tracked hours on Thunder Marketing's Project, with zero commits, zero PRs and zero AWS events. There is nothing in this dataset that shows what was produced. That is a data question first and a delivery question second, but it does need answering.	Zero infra events can mean no AWS identity is mapped for him, not that no infra work happened. If his GitHub account is not linked to his employee record, all of his code work would be invisible here. Work in unmerged branches, design, QA or support would also leave nothing in this data. Action: confirm his GitHub and AWS identity mappings before drawing any conclusion from this.
20	Anmol Music	needs-attention	0.40	115.4 tracked hours, all on a project named "Music", with no commits, no PRs and no AWS activity. Nothing in this engineering dataset describes the output, and the project name suggests the role may not be a coding one at all.	If this is a creative or media role, this developer-delivery dataset is the wrong lens and the low rank says nothing about the work. No AWS identity appears to be mapped, so any infrastructure work would be invisible. No GitHub activity at all, which could equally mean an unlinked account. Evidence confidence is 40 percent.
21	Srija Bandapally	needs-attention	0.40	16.1 tracked hours on Thunder Marketing's Project and nothing else in the period, with no commits, PRs or AWS events. The tracked time is real but too small to say anything about delivery.	There appear to be two employee records under this name, so hours and any output may be split across both. No mapped AWS identity, so any infrastructure work would not appear. 16 hours is a part-period presence, so this rank reflects a short window and not a full month.
22	Srija Bandapally	needs-attention	0.40	1.4 tracked hours on "Strategy & Financial Consultant" and nothing else. This is the thinnest record in the set and the project name suggests a non-engineering role.	This is almost certainly a duplicate record for the same person as employee 17, and should be merged. A consulting or finance role would not be expected to produce commits or AWS events, so this rank should not be read as a performance judgement. 1.4 hours is too little data to support any conclusion.

Not ranked — no evidence this period

Listed, never guessed. Ranking someone here would fabricate a comparison the data can't support.

Cursor Agent, Ruhani Allagh

literally no GitHub, Hubstaff or AWS evidence this period

What this data cannot see

No PRs were opened or reviewed by anyone this period, so code review, review comments and mentoring through review are completely absent from this data.
Production incident response that leaves no AWS write, such as reading logs, restarting something by hand on a box, a phone call or a Slack thread, is invisible.
Architecture and design decisions, planning sessions and technical direction leave no trace unless someone happens to commit a document.
Work in unmerged branches, local experiments and spikes that were abandoned do not appear at all.
Anyone without a mapped AWS identity shows infra.totalEvents of 0, which looks identical to doing no infrastructure work. Several people here are in that position.
Every Hubstaff project is unmapped to a repository, so tracked hours cannot be tied to any specific piece of code.
Cross-team help, unblocking colleagues, QA, customer support and recruiting or onboarding time are not measured anywhere.
No task tracker is connected, so nothing here confirms that what was built was what was asked for.