Cityfurnish's ATS Shows Only 20 Candidates. The Rest Never Surface.

Why Cityfurnish Is Adding Roles Now

Cityfurnish has multiple open roles on the Zero G Talent board (our platform, ingested directly from the company's ATS), with three posted in the past week. Every application passes through a Zoho Recruit pipeline that drops skills buried in bullet points and caps the ranked list at twenty names. The company is hiring across marketing, design, procurement, and founder-adjacent operations while its automated screen quietly decides which candidates a human ever sees.

Neerav Jain, Saurabh Gupta, and Vinit Jain founded the company in 2015. They spent a decade turning urban mobility into a subscription business: students, expatriates, and transient professionals rent furnishings, appliances, and fitness equipment for three to twelve months, with 72-hour delivery and maintenance included. What changed recently is the layer underneath. The platform now manages over 100,000 customers and 12,000 monthly rentals across Delhi NCR, Bengaluru, Mumbai, Pune, Hyderabad, and Chennai, per StartupIntros. A B2B channel serving student housing, co-living operators, and corporate clients opened a second revenue stream that demands different logistics, different data, and different people.

Capital followed traction. Cityfurnish raised $8.2 million across three rounds, StartupIntros reports, including a $3 million Series A in November 2022 backed by Y Combinator, Venture Highway, Global Founders Capital, and angels such as Paul Buchheit and Steve Chen. A $2.5 million debt-and-equity tranche closed the same month, Free Press Journal and Zee Business report. The most recent injection came January 2, 2025, via the Panasonic Ignition Accelerator Program, CB Insights found. By the end of 2022 the company had turned profitable and recorded more than 100 percent revenue growth, per Free Press Journal. GetLatka estimates annual revenue at roughly $28.8 million as of December 2024.

Headcount tells the same story. LinkedIn data puts the team at 188 in December 2024, GetLatka's figures show; Pitchbook counts 200 as of 2026. The StartupIntros profile, likely older, lists 90. Whatever the exact figure, the trajectory is upward. The company said it will use fresh capital to expand operations, advance its technology stack, strengthen its brand, and establish a presence in new cities by 2025, per Free Press Journal. Redseer values the Indian rental furniture and appliance market at Rs33,500 crore and projects an 11 percent compound annual growth rate through 2025, Livemint's data shows, driven by urban mobility, environmental awareness, and the normalization of subscription consumption.

That expansion requires hands. The live openings on Zero G Talent, all Gurugram-based, span multiple tracks. The Y Combinator "Work at a Startup" board lists the most technically demanding positions: Senior App Developer, Odoo Developer, and Data Analytics AI Specialist. Zero G Talent's board shows growth-engine roles: Brand Lead (INR 400K–800K/yr), Social Media Executive (Marketing) and Video Editor (INR 300K–350K/yr each), plus Video Editor Intern, UI UX Intern, Graphic Design Intern, Social Media & Influencer Marketing Intern, Procurement Intern, Founder's Office - Intern, and Human Resource Intern (stipends INR 10K–18K/month).

The concentration of creative and brand roles signals a shift: Cityfurnish is investing in acquisition and retention while it builds the automation layer that will let it scale without linear headcount growth. The Senior App Developer owns the mobile lifecycle — discovery, checkout, KYC, subscriptions, service requests — for a consumer product used daily across seven cities. The Odoo Developer extends the ERP core that drives supply chain, inventory, finance, and customer operations, bending a monolithic platform to a rental business's quirks. The Data Analytics AI Specialist translates messy operational data into churn models, demand forecasts, and pricing optimization that ship inside dashboards and automated alerts. The other roles are legitimate entry points into a funded, growing startup, but they don't carry the same technical bar.

How Cityfurnish's ATS Filters Candidates

Cityfurnish runs its hiring through Zoho Recruit at cityfurnish.zohorecruit.com, per the company's careers page. Every application, whether for a Social Media & Influencer Marketing Intern or a Brand Lead, passes through the same parsing and matching pipeline before a human sees it.

The parser is not Zoho's own. Zoho states directly that it uses DaXtra, a third-party engine, to extract name, contact information, skills, and work history from uploaded resumes, ApplyVita's review of vendor documentation shows. Job descriptions, by contrast, are processed by Zoho's in-house Zia parser. That split matters: resume-parsing behavior here is DaXtra's, not something Zoho built, and DaXtra's tolerances become the de facto rules for any candidate.

Zoho names four supported formats: DOC, PDF, RTF, and TXT, its product page lists. The attachment limit is 20 MB per record, Zoho's help documentation states. But format compliance is only the entry ticket. DaXtra struggles with complex layouts, tables, and multi-column designs, less so than enterprise ATS platforms, but enough that merged text in the experience section is a documented failure mode. In parser preview, bullet points collapse into a single paragraph, destroying bullet-level relevance scoring and recruiter readability, ResumeATS.net found. Job titles routinely parse incorrectly or merge with company names, which breaks role-level matching in recruiter search, ResumeATS.net found. Skills listed only inside bullet points often never reach the structured Skills field at all, ApplyVita reports.

ApplyVita's analysis of Zoho's documentation states: a skill mentioned only inside a bullet point may never reach that field, and if it does not, you are filtered out of a search you would have topped.

Zia's 0–100 match score — the ranking signal recruiters see — is calculated from the candidate's structured Skills and Qualifications fields, not from free-text resume prose matched against a job description, Zoho's help documentation cited by ApplyVita shows. Where a job opening lists required primary skills, only candidates who carry those exact skills in their skill set field appear in the match results, Zoho's documentation states. This is the closest thing to automatic exclusion in Zoho: a hard gate on a structured field, not a judgment about your resume.

The practical ceiling is tighter than most applicants assume. Zoho documents a maximum of twenty matching candidates displayed, its help center notes. On a well-subscribed role, ranking twenty-first is functionally the same as not appearing. The free tier of Zoho Recruit caps storage at 256 MB and does not include resume parsing; parsing unlocks in paid tiers starting around $25–30 per user monthly, US Chamber of Commerce and TechRepublic reviews indicate. Cityfurnish's volume, with multiple open roles across marketing, design, procurement, HR, and founder's office, confirms a paid deployment with parsing active.

Agency recruiters compound the effect. When a staffing firm processes your resume through Zoho Recruit, it enters a shared database searched across multiple client openings, Zoho's own marketing for staffing agencies explains. The same structured-field matching governs whether you surface, regardless of the end employer's size. That means a resume optimized for one Cityfurnish role must also carry the broad skill taxonomy that agency searches target.

Zoho publishes no formatting guidance, nothing on columns, tables, or graphics, ApplyVita's review found. That absence is not permission. The parser expects selectable text, a single column, and standard section headings. Primary keywords belong in the summary and first two experience entries. Tools, skills, and technologies need a clearly labeled Skills section. Title and company belong on distinct lines with no special formatting or symbols. DOCX tends to parse most reliably.

The system does not auto-reject. But the combination of a hard primary-skills filter, a twenty-result display cap, and a parser that routinely drops skills from bullet points means the filter is real; it just lives in the structured fields, not in a "reject" button. Candidates who treat the Skills field as optional disappear from the ranked list before a recruiter opens a single profile.

Insider Strategies for Beating the Screen

Jobscan's 2025 data, Novorésumé cites, indicates applicant-tracking systems reject up to 75 percent of résumés for parsing errors alone. At Cityfurnish, where Zoho Recruit powers the funnel, the same mechanics apply. The bot reads first. The recruiter reads second, if you're lucky. You get about six seconds of human attention once you clear the machine, multiple recruiter surveys cited by BOSS Magazine and LinkedIn show.

Format for the parser, not the portfolio

Complexity is the enemy. Tables, text boxes, multiple columns, graphics, charts, and colored backgrounds all scramble the extraction layer. When an ATS scans a table, it often reads left-to-right across rows, mashing dates, titles, and descriptions into a garbled string, OneTwoResume reports. Headers and footers are invisible to many parsers; put contact information in the body. Stick to a single-column, top-to-bottom layout with standard round bullets. Use conventional section headings — Work Experience, Education, Skills, Professional Summary — not creative labels like "My Journey" or "Where I've Been."

Font choice matters more than most candidates realize. If the system doesn't recognize the typeface, it defaults to squares or question marks, BOSS Magazine reports. Arial, Calibri, Helvetica, Roboto, Garamond, Georgia, and Times New Roman are safe. File format is a tactical decision: .docx remains the most compatible for online portals and LinkedIn Easy Apply. PDF works if the job posting explicitly requests it or you're emailing a person directly, but only if it's text-based, not a scanned image, multiple ATS guides advise.

Keywords: mirror, don't stuff

The mirror method works. Read the job description. Highlight every hard skill: "Project Management," "Python," "Financial Analysis," "Zoho Recruit," "React." If the posting says "Client Relations" and your résumé says "Customer Service," change it. Use their exact phrasing, including capitalization: "JavaScript" not "javascript." Include both acronym and spelled-out form: "Search Engine Optimization (SEO)." Modern ATS engines understand synonyms and context better than they did three years ago, ApplyVita finds, but exact matches still carry the heaviest weight.

Context is king. The system checks where a keyword sits. A skill listed under a 2018 role carries less weight than the same skill in your most recent position or in a dedicated Skills section, ApplyVita and OneTwoResume find. If you haven't used a technology in ten years, the bot knows. A dedicated Skills section, clearly categorized (Technical Skills, Certifications, Tools), is non-negotiable for parse accuracy.

BOSS Magazine summarizes: the bot gets you in the pile; the numbers get you the interview.

Quantify everything humans will scan

Robots look for nouns. Humans look for numbers. "Managed a team" disappears. "Managed a team of 12 and increased sales by 20 percent" lands. BOSS Magazine reports that résumés packed with hard metrics receive roughly 60 percent more interview invitations, yet fewer than one in five executive documents include them. Every bullet should follow the pattern: action verb + scope + quantified result. "Cut costs by 12 percent across APAC." "Reduced churn from 8 percent to 3 percent in six months." "Shipped 14 features on a 6-person squad."

White space is not wasted space. A wall of keyword-stuffed text fails the six-second human scan. Recruiters need clear headings, breathing room, and a visual hierarchy that guides the eye to your strongest evidence. BetaNews reports that over 40% of recruiters now screen résumés on phones; dense blocks become unreadable on mobile.

AI as co-pilot, not author

BOSS Magazine reports that 45 percent of job seekers already use AI for polish, and 90 percent of hiring managers say that's acceptable, a 2024 Forbes/Sago study found. But 41 percent of recruiters say they can detect AI-heavy résumés, CareerBuilder's 2025 survey shows, and CareerBuilder's white paper shows they struggle to spot human-edited AI text. The line is thin. Use AI to tighten verbs, surface missing keywords, and catch typos, then fact-check every number against your own dashboards or P&L reports. Generative models fabricate details.

A practical loop: draft in plain language with real scope and results. Run the AI enhancer. Accept stronger verbs and clearer structure; reject any invented metrics. Compare the revised bullet to source data before saving. Tools like Teal or Rezi can scan a job description and flag the five to seven most-repeated phrases — "digital transformation," "P&L ownership," "cross-functional leadership" — in seconds, BOSS Magazine's hands-on testing shows. Insert those terms where they're accurate for you. Re-scan until the ATS score climbs above 85. BOSS Magazine reports that users following this workflow cut edit time from 45 minutes to under 12 minutes per application, Teal's November 2025 usage data shows.

Test before you submit

Three free checks take two minutes. Copy your résumé text into a plain-text editor; if it reads cleanly, the parser likely handles it. Upload to LinkedIn's résumé builder; it mimics ATS parsing. Run a free ATS scanner (Jobscan, ResumeVera, or the builder's native checker) and aim for a parsing score above 90. Then hand the PDF to a mentor or former colleague. Ask: "Does this read like how I lead in real life?" A five-minute peer review catches tone issues and factual drift that algorithms miss, BOSS Magazine notes.

Tailor every application

OneTwoResume reports that 73 percent of hiring managers prioritize candidates who match at least 80 percent of the technical skills listed in the posting, but they interview the ones with strong soft skills. Keep a master résumé with every quantified win. For each role, clone the file, run the keyword scan, insert the top phrases where they're true, reorder bullets to lead with relevant experience, and adjust the summary to reflect the specific title. Ten minutes. Label the export "Cityfurnish-BrandLead-Dec25.pdf" so you track which version went where.

The hidden screen is beatable. It rewards discipline over creativity, specificity over volume, and verification over trust. The candidates who clear it aren't gaming the system; they're speaking its language well enough that the human on the other side sees the evidence, not the formatting.

How AI and Automation Power Cityfurnish

Cityfurnish's technology stack scores 86 out of 100 on modernity, placing it in the "cutting-edge AI adopter" tier, a figure derived from 49 tracked technologies across 18 recent job postings, per Alion. The stack runs on Python (17% of postings), JavaScript (11%), SQL (11%), and Node.js (6%), with NumPy and Pandas each appearing in 11% of listings. PostgreSQL leads databases at 11%, followed by MySQL at 6%. Frontend work centers on Next.js (6%), Tailwind CSS, Bootstrap, and Vite. Analytics lean on Power BI and Tableau, both at 11%.

The AI layer is where the rental model meets frontier tech. ChatGPT, Claude, and generic LLM references each show up in 11% of postings. Machine Learning, SLI/SLO/SLA, and Prompt Engineering also sit at 11%. CI/CD appears in 6%. This isn't a side experiment. In the last 30 days, an AI workforce built with SagePilot handled more than 275,000 lifecycle touchpoints across WhatsApp, email, and voice, SagePilot's case study shows. Subscription renewals alone drove 38,000+ conversations. Abandoned carts were re-engaged 21,000 times. Referral flows fired 18,000+ times after invoices and deliveries. Overdue outreach now runs as a seven-band ladder from day 0 to day 300, with voice agents joining at the exact bands where a call outperforms a message.

SagePilot's case study states: the AI workforce runs the rental lifecycle end to end, with 275,000+ lifecycle touchpoints handled in the last 30 days.

Two named agents illustrate the depth. Sneha intercepts failed checkout payments within the hour, explains the issue, holds the delivery slot, and sends a fresh payment link on WhatsApp before the customer moves on. Anjali holds the inbound line across WhatsApp, Instagram, Messenger, and web chat, verifying invoices and order state before answering so "I already paid" becomes a clear explanation instead of an escalation to accounts. Humans see only the cases that need judgment, pre-briefed with the full story. Routine calls and reminders are no longer placed by hand.

The rollout happened in phases. Phase 1 put Anjali live on chat, trained on rental tenures, KYC requirements, billing, and refund policy. Phase 2 added proactive outreach triggered by order and billing events: signup, KYC, delivery, renewal, referral, overdue. Phase 3 deployed eight voice agents for welcome calls, KYC nudges, delivery scheduling, failed-order rescues, retention, and the collections ladder. Today the workforce runs the full lifecycle. No new technology has appeared in Cityfurnish job postings in the last six months, Alion found, suggesting the stack has stabilized around this AI core.

The creative side of the house runs on Adobe After Effects and Photoshop (each 39% of postings), Canva (33%), and Premiere Pro (28%). Figma appears at 6%, Alion's analysis of marketing-team postings shows. Marketing roles, currently 41% of hiring focus, Alion's data shows, depend on this toolchain. The Zero G Talent board lists openings for Video Editor, Video Editor Intern, Graphic Design Intern, Social Media Executive (Marketing), and Social Media & Influencer Marketing Intern, all Gurugram-based. The founders have built a company where the rental operational complexity — KYC stalls, failed payments, delivery scheduling, monthly invoices, renewals, buybacks, ownership transfers, and an overdue clock that can reach day 300 — is managed by code and models, not headcount alone, SagePilot explains. That's why the open roles skew toward data science, ML engineering, and product roles that can extend the AI workforce rather than just staff the manual work it replaced.

How Candidates and Recruiters Are Responding

The numbers on Cityfurnish's employer profile tell a story of mediocrity. AmbitionBox aggregates over 66 employee reviews into an overall rating of 3.3 out of 5, a score the site itself summarizes as "most employees have an average experience working at Cityfurnish."

Recruiters at these companies face a mirror-image problem. LegalLancerIndia's analysis of the Indian startup hiring landscape identifies three persistent failure modes: high costs, slow recruitment cycles, and mismatched talent. When a single role draws hundreds of applications, common for marketing, design, and founder's-office positions like the ones Cityfurnish has posted recently, the only way to move fast is to automate the first pass. But automation calibrated for speed inevitably discards edge cases: the candidate who built a relevant project outside a standard job title, the career switcher whose transferable skills don't map to the keyword list, the applicant who formatted a clean PDF that the parser reads as garbage text.

The market is reacting. LinkedIn's 2025 outlook on India's startup hiring surge notes the rise of AI tools on both sides of the table: recruiters deploying smarter parsing, candidates using AI to reverse-engineer job descriptions and rewrite resumes for each application. RKHRM's 2025 playbook for Indian HR teams advises: audit your ATS rules quarterly, keep a human in the loop for borderline profiles, and publish the actual skills you screen for rather than hiding them behind generic requirements.

For Cityfurnish specifically, the past week's board activity shows three new roles: Social Media & Influencer Marketing Intern, Social Media Executive (Marketing), and Procurement Intern, joining a pipeline that already includes Video Editor, Brand Lead, and Founder's Office - Intern. That volume, concentrated in Gurugram, means the parsing rules are under pressure.

Recruiters, meanwhile, are learning that the filter they set today becomes the talent pool they inherit tomorrow. Glassdoor's long-term research, drawn from over 2 million anonymous reviews across nine years of "Best Places to Work" data, found that culture and mission alignment — not pay — drive satisfaction, and that satisfied employees causally outperform the market, Glassdoor CEO Robert Hohman said on CNBC. The implication for a company running a tight keyword screen is blunt: if your ATS optimizes for keyword density, you hire keyword optimizers. If you want builders who align with the mission, the screen has to let through the language of builders — project links, scope descriptions, outcome metrics — not just the right nouns.

The tension is visible in the Trustpilot footprint, too. Twenty-three customer reviews sit alongside the employee reviews; one Reddit user documented a billing dispute where "Cityfurnish deducted double the agreed amount from my account every month." That kind of operational friction, visible to candidates who research the company before applying, feeds into the same perception loop: a company that automates its customer interactions poorly may also automate its hiring poorly. The candidates who persist, tailoring resumes, following up on LinkedIn, and reaching out to the founders directly, are effectively stress-testing the company's human layer. The ones who don't never reach the structured Skills field that Zia reads. The parser never sees them. The recruiter never sees them. The role stays open.

Working in frontier tech? Zero G Talent tracks the openings: see every open Cityfurnish role, browse frontier tech jobs, the companies hiring, and the people building the field.