What Model ML Builds — and Why It's Staffing Up
Model ML, the AI workspace that top financial firms now treat as strategic infrastructure, is hiring for 15 roles across engineering, product, and go-to-market, and its recruiters filter for proven project impact, visible code, and warm referrals above all. The hiring wave signals something larger than headcount growth: a template for how AI-native companies are staffing up in 2025.
Chaz and Arnie Englander founded Model ML in 2023 after selling Fancy, a last-mile delivery company, to Gopuff, and Llama, a marketplace, to Hygglo. They started Model ML as an internal tool to guide their own investments, then spun it into a company after their network asked for access. A $12 million round led by Y Combinator and LocalGlobe closed in February 2025.
The platform sits on a firm's own infrastructure and connects to every data source the team touches: emails, files, CRM, databases. Its interface mirrors Google Drive, complete with spreadsheet, document, and presentation clones so information never leaves the workspace. Users prompt in natural language to generate company profiles, 100-page memos, or comparable company analysis charts that link every data point to its source. Chaz Englander, the CEO, described it to Fortune as "effectively a tabular view on top of an agentic system." Templates let analysts set up recurring work (comps, earnings summaries, trend analysis, market mapping) once and rerun it.
The product targets investment banks, private equity firms, and family offices. Englander told Fortune the company has roughly 40 customers, including many of the largest financial organizations globally, though he declined to name them. "This is going to play such a fundamental part in their overall success as a business that people (and rightfully so) are mindful about how they're articulating themselves, what they're automating, and what they're not automating," he said.
The hiring push arrives as financial institutions race to deploy AI. Bloomberg Intelligence reported last month that Wall Street could shed as many as 200,000 jobs over the coming years, with CIOs surveyed expecting a 3 percent workforce reduction on average and one in four predicting cuts of 5 to 10 percent. Routine, repetitive tasks are first on the chopping block. Model ML's pitch — automating the grunt work of due diligence while keeping humans in the loop — lands in that exact gap.
Competition is thick. Kavout, Finomial, Alphasense, and Samaya operate in adjacent territory. OpenAI, Google, and Anthropic have all released deep-research agents that compile complex reports. Finster AI, another AI-native platform for investment banks and asset managers, closed a $15 million combined Seed and Series A in November 2025. Model ML's differentiation rests on its workspace model — data never leaves the firm's environment — and its focus on financial-services workflows rather than general research.
The company employs 80 people in London. First-party board data shows 15 salaried roles open with a typical band of $74,000 to $230,000 (median $140,000). Recent postings include a GTM Lead in New York at $200,000–$400,000, a Head of IT & Compliance in New York at $150,000–$250,000, and engineering roles in London spanning applied AI, backend, and full-stack data science. Traffic to the site hit 7,700 visits last month, up 28.6 percent, with 77.5 percent from the U.S. and 22.5 percent from the U.K.
Nearly one in five seats is empty.
The 15 Open Roles: Where They Sit and What They Demand
Model ML's current hiring wave spans 15 salaried positions, concentrated heavily in London with a strategic foothold in New York and a distributed growth role across Asia-Pacific. The board's live data shows a salary band of $74K–$230K with a $140K median, but the spread is wide: an Operations role open to new graduates starts at $35K, while the New York GTM Lead tops out at $400K. That range reflects a company building both depth and breadth at once: senior ICs in applied AI and infrastructure, a full product growth function, a creative studio in-house, and a go-to-market engine that now stretches from Dubai to Singapore.
Engineering and applied research carry the most headcount. Six of the 15 roles sit here: Applied AI Engineer (London, $100K–$200K, 6+ years), Senior Backend Engineer (London, $120K–$180K, 6+ years), Full-Stack Engineer (London, $80K–$140K, 6+ years), Infrastructure/Security/DevOps (London, $80K–$140K, 6+ years), Full Stack Data Scientist (London, $100K–$160K, 6+ years), and Head of IT & Compliance (New York, $150K–$250K, 6+ years). Every technical role demands six-plus years of experience: no junior hires, no "high potential" labels. The message is clear: Model ML needs engineers who have shipped production systems in regulated or high-stakes environments, not researchers looking for their first industry role.
| Role | Function | Location | Salary Band (USD) | Min. Experience |
|---|---|---|---|---|
| Applied AI Engineer | Engineering | London | $100K–$200K | 6+ years |
| Senior Backend Engineer | Engineering | London | $120K–$180K | 6+ years |
| Full-Stack Engineer | Engineering | London | $80K–$140K | 6+ years |
| Infrastructure/Security/DevOps | Engineering | London | $80K–$140K | 6+ years |
| Full Stack Data Scientist | Engineering | London | $100K–$160K | 6+ years |
| Head of IT & Compliance | Engineering | New York | $150K–$250K | 6+ years |
| Product Growth | Product | New York / Remote London | $70K–$130K | 6+ years |
| Data, Strategy, & Ops | Product | London | £90K–£130K (~$115K–$165K) | 6+ years |
| GTM Lead | Go-to-Market | New York | $200K–$400K | 3+ years |
| Growth | Go-to-Market | SG / HK / IN / US / Remote | $120K–$200K | 6+ years |
| Marketing Lead | Go-to-Market | London | $80K–$130K | 3+ years |
| Design Lead | Design | London | $80K–$140K | 3+ years |
| Brand Designer | Design | London | $80K–$120K | 3+ years |
| Content Strategist | Design | London | $80K–$120K | 3+ years |
| Operations | Operations | London | $35K–$50K | Any (new grads ok) |
Product and data strategy account for two roles, both in London bar the Product Growth position split between New York and a remote London option. Data, Strategy, & Ops (£90K–£130K, 6+ years) signals a team building internal instrumentation and customer-facing analytics, critical for a product that automates financial due diligence and needs to prove ROI to investment-bank clients. Product Growth ($70K–$130K, 6+ years) sits at the intersection of PLG motion and enterprise sales support, a hybrid that only exists because Model ML's inbound flywheel (content-driven, per the founders) now needs systematic conversion infrastructure.
Go-to-market is where the geographic spread widens. The GTM Lead in New York ($200K–$400K, 3+ years) carries the widest band in the entire slate: variable-heavy, enterprise-quota territory. Growth spans Singapore, Hong Kong, India, and the US on a single requisition ($120K–$200K, 6+ years), reflecting a deliberate APAC push rather than opportunistic hiring. Marketing Lead in London ($80K–$130K, 3+ years) rounds out the commercial trio. Notably, all three GTM roles ask for three-plus years, not six, which is the only function where the experience floor drops, suggesting Model ML will train the right commercial instincts if the raw horsepower is there.
Design and content get three dedicated roles, all London-based, all at the 3+ year threshold: Design Lead ($80K–$140K), Brand Designer ($80K–$120K), Content Strategist ($80K–$120K). This isn't a branding afterthought; the company's own careers page states that "most of our inbound today comes from the content we publish" and they are "becoming a company where the story, the product narrative, and the visual identity attract customers before any formal sales motion." The creative hires are the engine of that flywheel.
Operations sits alone at the entry level: London, $35K–$50K, new graduates welcome. It's the only role without a seniority gate, and the only one paid in a band that doesn't clear six figures. In a 15-role slate dominated by six-year minimums, this is the deliberate exception: a pipeline bet, not a cost center.
Geographically, 11 of 15 roles anchor in London. Three sit in New York. One is explicitly distributed across APAC and the US. The board's first-party data confirms this concentration: the latest additions, including GTM Lead (New York), the compliance lead (New York), Growth (APAC/US remote), Applied AI Engineer (London), Senior Backend Engineer (London), and Full Stack Data Scientist (London), reinforce the twin-hub model with a London-heavy engineering core and a New York commercial tip. For candidates, the implication is straightforward: London is where the product gets built; New York is where it gets sold; APAC is where the next revenue frontier opens.
What Model ML's Hiring Velocity Signals About Screening Intensity
Model ML's shipping cadence sets the bar for what its recruiters screen for. In a single month this summer the company released an agent framework, Excel error-checking agents, a Teams plug-in, MCP support in both directions, Snowflake and LSEG integrations, a public financial-services benchmark (the Model ML Composite), and new offices in Madrid, Paris, Singapore, and Tokyo. That velocity — documented in the company's LinkedIn posts — means the recruiter screen filters for engineers who can match that pace: production-grade code, documented impact, and the ability to move from ambiguous problem to shipped artifact without hand-holding.
Industry data shows that 70–80% of applicants are rejected at the recruiter screen across AI companies, not for lack of skill but for lack of clarity, communication, or preparation. The recruiter's write-up becomes the narrative blueprint for the entire interview loop; every subsequent interviewer reads it before they meet you. At Model ML, where the engineering team ships weekly, the screen is a de-risking filter: can this candidate communicate technical work in structured, concise terms? Do they connect their ML work to business outcomes? Can they drive problems end-to-end, not just execute assigned tickets? The distinction between candidates who built models in notebooks and those who shipped models into production environments matters acutely here.
Referral weight follows the industry pattern: warm introductions from current employees or trusted network nodes bypass the cold inbound pile. Model ML's recruiters act as strategic hiring partners, shaping the search, not just the candidate flow, and identifying early mismatches in production experience, tool alignment, compensation expectations, or interest in the company's problem set.
Portfolio depth matters more than credentials. A GitHub repository showing a model shipped to production, with documentation on data pipelines, monitoring, and rollback procedures, carries more weight than a publication list. Recruiters want to see the diagnostic process: how you identified the problem, what you changed, what you watched to know if it worked. That four-step structure (definition, diagnostic process, decision, production consequence) separates a practitioner from a textbook reader. The candidate who clears the screen is the one who can walk into an ambiguous problem, name what's uncertain, propose a reasonable path, and explain what they'd watch for to know if it was working.
Candidate Playbook: How to Tailor Your Application
Model ML's recruiters filter for three signals above all: proven project impact, visible code or portfolio evidence, and a warm introduction. The company's August LinkedIn post put it plainly — they want "the hardest working, fastest moving team in AI" — and the shipping cadence backs that up. That velocity shapes what the screen rewards.
Referral strategy: map the team, then ask for the intro
Cold applications land in a queue; referred candidates land in a conversation. The research names several engineers and product leads who have shipped visible features recently: James (Head of AI, PhD Theoretical Physics, Imperial College London) led the Composite benchmark and MCP work; Diogo (Product Manager, former Bain consultant) owns the "Sites" HTML dashboard product. Each is a credible referral source because they own a shipping surface. Other recent shippers include Edward (MCP server, Snowflake connector), Simone and Ahmet (scheduled and event-based agents), Isaac (Teams plug-in), Somesh (AutoCheck for Excel), and the enterprise deal team (Raghav, Milan, Tejas, Chester) behind the PwC and Deloitte contracts.
Start by following the company page and the individuals above on LinkedIn. Comment on their launch posts with a specific technical observation (not "congrats") then request a 15-minute coffee chat referencing that observation. Example: "Saw Edward's post on the Custom MCP server. We're evaluating MCP vs. function-calling for a similar agent loop at my current shop; would value 15 minutes on how you decided on the harness boundary." That signals you read the code, not just the announcement.
If you lack a direct connection, target the "growth" and GTM roles listed on the Zero G Talent board (GTM Lead New York $200–400k, Growth SG/HK/IN/US/Remote $120–200k). Those hires sit beside the engineering pod and often run referral bonuses. A warm intro from a GTM hire carries weight because they vet for "customer obsession", which is the other half of the company's stated identity.
Resume keywords that survive the screen
Model ML's stack is Python-heavy on the agent harness and backend, TypeScript/React on the frontend "Sites" dashboards, and MCP/JSON-RPC for integrations. The board lists Applied AI Engineer (London £100–200k), Senior Backend Engineer (London £120–180k), and Full Stack Data Scientist (London £100–160k). Keywords that map to those reqs:
- Agent harness / tool-calling loop / ReAct pattern: James's LinkedIn explainer defines the harness as "the layer that determines whether an agent delivers client-ready output or stops halfway." Use that phrase.
- MCP (Model Context Protocol) server / client implementation: they ship both directions; the July 22 post notes "MCP in both directions: bringing external tools and data into Model ML, and making Model ML's agents available elsewhere."
- RAG over financial filings / multi-document reasoning: the Composite benchmark tests "frontier multi-document intelligence" across data rooms, cross-filings, research memos.
- Quantitative research / regression / filing reconciliation: Opus 5 evaluation cites "class-leading quantitative research: multi-company screens, regressions, filing reconciliations."
- ISO 27001 / SOC 2 Type II / GDPR compliance: the marketing page leads with these; enterprise deals (PwC, Deloitte, FT Partners) require them.
- Model routing / cost-aware inference: "No single model wins on both price and performance… Model ML routes each task to whatever model is best suited to it."
Place these in a "Technical Highlights" section under each role, not in a keyword cloud. One bullet per project: "Built MCP server exposing internal pricing engine to Claude; cut analyst lookup time 60%." That beats a list.
Portfolio projects that mirror Model ML's work
The screen favors artifacts over descriptions. Three project archetypes map to open roles:
Agent harness demo: A minimal loop: user request → planner → tool calls (search, code exec, file write) → verifier → final artifact. Host on GitHub with a README that explains the harness boundary decisions (timeout policy, retry logic, context window management). Cite James's harness definition in the README, which shows you've read their engineering blog.
MCP integration: Wrap a real financial data source (SEC EDGAR, Yahoo Finance, or a synthetic deal room) as an MCP server. Demonstrate a Claude Desktop or Cursor session calling it. The July 22 post explicitly invites "custom MCP servers without relying on a predefined integration list."
Finance workflow clone: Replicate one Model ML feature: AutoCheck for Excel (flag hardcodes, broken links, inconsistencies), or a "Sites" dashboard that drills from summary → model → source contract in three clicks (Diogo's demo). Use Streamlit or a React + FastAPI stack. Deploy to a public URL; recruiters click links.
If you lack finance domain data, say so in the README and show how you'd swap the data layer. The company hires generalist engineers who learn the domain fast; Diogo came from Bain, not a bank.
Tailor the application packet
The contact form at modelml.com/contact-us asks: "How did you hear about us?" (options include LinkedIn, Friend or Colleague, Event or Webinar) and "Did you use AI to evaluate us?" Answer honestly: they build AI eval tools; they'll spot generated fluff. Attach a one-page addendum: "Why Model ML": three bullets tying your last project to a specific launch they shipped this quarter. Reference the Composite benchmark, the PwC/Deloitte/FT Partners logos, or the new office geography if you're in Singapore, Tokyo, Paris, or Madrid. That signals you track the company, not just the role.
The screen is narrow by design. A referral plus a live repo that speaks their vocabulary gets you past it. Everything else is noise.
Market Context: What This Hiring Wave Signals
Model ML's €65 million Series A and 15-role hiring sprint is not an isolated event: it is a data point in a structural shift that has been building since the Federal Reserve cut rates by half a point in mid-September 2024, bringing the benchmark to 4.75–5 percent. CompTIA's Tim Herbert called the move "the greenlight to move forward in addressing their tech talent needs," and the numbers bear that out: U.S. tech unemployment sits at 2.5 percent as of October 2024, while AI-specific vacancies rose 5.6 percent from Q1 to Q2 2024 and 31.5 percent year-over-year. The median AI salary hit $157,196 in Q2 2024, but that figure masks a widening split.
In London, where Model ML is headquartered, the divergence is starker. Santa Monica Talent's 2026 Salary Benchmark reports a two-tier market emerging at "unprecedented speed": AI-specific roles command £250,000-plus while the broader tech market stagnates. Burns Sheehan's London Technology Salary Guide 2026 shows AI hiring up 130 percent year-on-year, concentrated in Series B–C companies, which is exactly the stage Model ML is entering. AI product managers now earn 25–35 percent more than their non-AI peers. Model ML's own bands (GTM Lead at $200–400k, Applied AI Engineer at $100–200k) reflect this premium.
The talent pool is expanding to meet that demand. OCI Insights tracked a 9.3 percent increase in U.S. AI professionals from Q2 to Q3 2024 increased 5.7 percent in academia and 10.1 percent in industry, with California (33,000), Washington (12,000), and New York (5,000) leading concentration. Yet 171,000 software development, cybersecurity, and data analytics roles remain unfilled nationwide. Hirewell data shows August 2024 starts up 50 percent year-over-year, September up 95 percent, and tech hiring up 150 percent, a 43 percent Q3-over-Q3 jump. The acceleration is broad but deepest in commercialization-facing functions.
That commercialization pivot is the through-line. Anthropic hired Instagram co-founder Mike Krieger after launching its iPhone app. Mistral brought in a North America GM with chief revenue officer experience. Perplexity added advisers from Uber, Android, and Bing. Model ML's open roles (GTM Lead, Growth, Applied AI Engineer) mirror this pattern: startups flush with Series A capital are racing to turn core models into revenue-generating products. The Ilya Sutskever departure from OpenAI underscores the tension between safety research and shipping velocity, but the market signal is clear: capital follows product traction.
A counter-current bears watching. A King's College London study from October 2025 found AI reshaping the UK labor market with "significant declines in employment and wages" at high-paying firms and professional occupations, while lower-paid sectors held steady. If that trend persists, the two-tier salary split could harden into a two-tier employment structure: AI-native roles absorbing premium compensation while adjacent technical roles face compression.
For candidates, the implication is direct: the premium attaches to demonstrable product impact in AI-native environments, not generalist engineering pedigree. Model ML's screen — shipped code, documented impact, a warm handshake — is the template. The companies raising Series A and B rounds in the next 12 months will run the same playbook. The hiring wave is not peaking; it is specializing.
The Englander brothers built their last companies to exit. This one, they're staffing to stay.
Working in frontier tech? Zero G Talent tracks the openings: see every open Model ML role, browse frontier tech jobs, the companies hiring, and the people building the field.