Careers at Autonomous Technologies Group: Teams, Pay and How to Get Hired

Who Gets Hired at Autonomous Technologies Group

Two founders who sold a GPU cloud company for a nine-figure exit, DataGod's Wikipedia background reports, are now building an applied AI lab for financial markets — and they're hiring the early team that will shape it.

Autonomous Technologies Group launched in 2025 from Y Combinator's fall batch with eight people, a nine-figure prior exit, and backing from Garry Tan, BoxGroup, and a quant fund that has backed Plaid, Ramp, and Stripe. The product, called Autonomous, is a next-generation financial advisor powered by frontier reasoning systems. The hiring plan is deliberate: roughly half the open roles sit in AI and research engineering, the other half in sales and client-facing work, all concentrated in New York and San Francisco with remote flexibility for certain positions.

The engineering side breaks into several distinct tracks. Data platform engineers, the role with the steepest experience bar at eight-plus years, own the data lake that ingests financial and market datasets from dozens of vendors. The job is to model that data so it becomes queryable, versioned, and trustworthy for both human researchers and AI agents. Financial-data fluency (market data, security identifiers, corporate actions, quantitative research) is explicitly preferred. Backend and full-stack engineers (three-plus years each) build the production systems that serve the advisor product. Research engineering (one-plus years) and quantitative systems (three-plus years) sit closer to the modeling layer, while a dedicated iOS role (three-plus years) handles the mobile surface. Research scientists need six-plus years and a track record of original work on reasoning in complex environments.

Design and client-facing roles round out the picture. A product designer with three-plus years owns the interface between sophisticated financial users and an AI-native system. The client director role, five-plus years, owns high-net-worth relationships end to end: sourcing, onboarding, adoption, CRM build-out, and feeding product insights back to engineering.

Across every technical role, the baseline includes comfort with an AI-native workflow. The team uses coding agents daily and builds data systems designed to be machine-consumable from day one. Python and SQL dominate the stack; AI agents appear in half the postings analyzed. The company describes the environment as "high agency, talent dense, zero bureaucracy" — a small, elite team building from first principles in a massive market.

The through-line is financial-domain depth paired with AI-systems fluency. ATG isn't hiring generalists who can learn finance later. It's hiring engineers and researchers who already speak the language of market data, and client leaders who already navigate high-net-worth relationships — people who can move fast because they don't need the domain translated for them.

What Autonomous Technologies Group Pays

ATG compensates like a well-funded early-stage lab that needs senior talent yesterday. The company's nine posted roles, all salaried, span a board-wide band of roughly $143,000 to $300,000 per year with a median around $260,000. Every role sits in New York City, San Francisco, or offers U.S. remote flexibility, and the bands reflect the premium the market places on applied AI researchers and engineers who can ship financial-grade systems.

The top of the market ($300,000), as Zero G Talent's data shows, appears on three distinct tracks: Client Director, Quantitative Systems, and Research Scientist. That convergence signals where ATG puts its highest leverage: revenue-facing leadership, the quant stack that powers the core product, and the research scientists advancing the "frontier reasoning systems" the lab describes on its site. The Data Platform role carries the widest spread ($125,000 range), which often indicates a role that could be filled by a senior individual contributor or a lead-level hire depending on the candidate's depth with large-scale financial data pipelines.

Equity details are not published in any of the public postings or the board data. For a Y Combinator F25 company with an eight-person team, the standard early-stage grant for technical hires typically falls in the 0.1–0.5% range, but candidates should treat that as market context, not a company figure. The same applies to benefits, refresh cadence, and performance bonuses; none appear in the listed sources. Ask directly in the first conversation.

Compared to public comp data for NYC AI labs, ATG's bands sit at or above the 75th percentile for individual-contributor engineering roles and are competitive with quant-focused shops that recruit from the same talent pool. The Product Designer band is the outlier at $115,000–$125,000, below market for a senior designer in New York, which may reflect a narrower scope or an earlier-stage title calibration. If you're evaluating an offer, the missing equity and benefits picture is the gap to close before you can compare total compensation fairly.

How the Hiring Process Works at Autonomous Technologies Group

ATG is an eight-person team hiring for seven open roles as of September 2026. That ratio, nearly one open seat per existing employee, means every hire changes the team's composition materially. The process reflects that weight: there is no high-volume funnel, no automated screening layer, and no standardized "university recruiting" track. Candidates enter through the careers page at atg.science/careers, the Y Combinator jobs board, startup.jobs, or Alion, and each application is reviewed by the technical leads who will work with the new hire.

Entry Points and Initial Screening

The company lists roles with explicit experience floors: Research Scientist (6+ years), the Data Platform MTS role (8+ years), Quantitative Systems MTS (3+ years), Member of Technical Staff (Research Engineering) (1+ years), Member of Technical Staff (Full Stack) (3+ years), the Backend MTS role (3+ years), the iOS MTS role (3+ years), Product Designer (3+ years), and Client Director (5+ years). These are not aspirational ranges; they are the minimum credible background for the work. A resume that doesn't meet the floor for the target role is unlikely to advance. The careers page states the team seeks "curious minds from a wide range of disciplines and backgrounds," but the role specs make clear that curiosity must be backed by demonstrated depth in the relevant domain: distributed systems for Data Platform, quantitative finance or ML systems for Quantitative Systems, iOS platform internals for the iOS role, and so on.

What the Public Record Shows

Glassdoor hosts 14 interview questions and 14 interview reviews posted anonymously by candidates who interviewed at "Autonomous" (the company's product name and common shorthand). The research available to this guide does not include the text of those questions or the review content, so the specific topics (coding challenges, system-design prompts, research discussions, domain-knowledge probes) cannot be reproduced here. What the volume suggests is a multi-stage process that generates enough candidate throughput to produce 14 public data points while the team remains under ten people. For a lab of this size, that typically means a phone screen with a founder or technical lead, a technical work session (live coding, system design, or paper review depending on role), and a final conversation with the broader team to assess collaboration fit.

Building a Strong Application

The strongest applications mirror the company's stated mission: "teach Autonomous to 1. deeply understand the market, 2. deeply understand an individual and their financial life, and 3. advance its representational capabilities." Candidates who show prior work at the intersection of large-scale reasoning systems and financial domain knowledge (published research, open-source contributions to data platforms or ML infrastructure, production experience with high-stakes numerical systems) signal alignment without needing to explain it. A cover letter that references a specific technical challenge from the company's blog or product description (e.g., the representation of long-horizon financial planning, the integration of unstructured market data into a reasoning loop) carries more weight than a generic statement of interest.

Practical Guidance

• Target one role. The experience floors are real; applying to multiple disparate roles signals indecision.
• Show the artifact. A link to a GitHub repo, a paper, a design doc, or a shipped feature beats a list of keywords.
• Prepare for depth, not breadth. With a team this small, interviewers will probe the limits of your claimed expertise rather than survey a wide surface.
• Expect speed. Eight people can move from screen to offer in days if alignment is clear; they can also stall if the signal is mixed. Follow up once, concisely, after each stage.

The hiring process is the company's first product decision for each candidate. Treat it like a design review: come with a thesis, evidence, and questions about the constraints.

Where the Work Happens at Autonomous Technologies Group

ATG operates from two primary physical hubs: New York City, which serves as its headquarters, and San Francisco, its second major engineering center. The company's own website describes it as "an applied AI research lab in New York City and San Francisco," and its public launch coverage confirms the bi-coastal footprint — TechStartups noted the "New York- and San Francisco-based AI startup has emerged from stealth with $15 million in pre-seed funding." This dual-location model is not incidental; it maps directly to the talent pools ATG draws from: quantitative finance and wealth-management infrastructure in New York, and frontier AI research and systems engineering in the Bay Area.

The New York office anchors the company's product and research direction. Multiple current postings on the Zero G Talent board list "New York City" as the sole or primary location: Research Scientist, Member of Technical Staff (Backend), and Member of Technical Staff (Research Engineering) are all NYC-only roles. The Client Director role, a senior, client-facing position with a $200k–$300k band, also includes New York as a base option alongside San Francisco and remote. For candidates, this signals that the core R&D loop, particularly the quantitative systems and research-engineering work that differentiates ATG's "autonomous AI financial advisor," runs through the Manhattan headquarters. The company's self-description as "headquartered in NYC" on atg.science reinforces that strategic and leadership functions sit there.

San Francisco functions as a co-equal engineering hub rather than a satellite. The Quantitative Systems MTS role lists "New York City / San Francisco, CA, US" as interchangeable locations, and the Data Platform role offers "New York City / San Francisco / Remote (US)." The Client Director role likewise treats San Francisco as a peer location to New York. This parity suggests ATG has built meaningful technical capacity in the Bay Area — likely to tap into the concentration of large-scale systems engineers and ML infrastructure talent that clusters around the major labs and hyperscalers. Candidates interviewing for quantitative-systems or data-platform work should expect the option to anchor in either city, with the team split across coasts.

Remote eligibility appears role-specific rather than company-wide. The board data shows three postings that explicitly include "Remote (US)" as an option: Client Director, the Data Platform MTS role, and the Quantitative Systems role (which lists only the two cities). The backend, research-engineering, and research-scientist roles do not list remote. This pattern aligns with an early-stage lab that requires tight in-person collaboration for core research and low-latency infrastructure work, while allowing flexibility for platform engineering and client-facing leadership. Candidates should treat the posting's location line as the authoritative signal — if remote isn't listed, assume on-site is required.

What the research does not disclose is the physical nature of either workspace: square footage, lab build-out, compute cluster access, or whether ATG occupies its own floor, a sublet, or a co-working build-out. No first-party blog posts or press coverage describe the office environment, and the company's engineering blog (autonomoustech.ca) focuses on technical frameworks rather than facilities. For a candidate, the absence of detail means the interview process is the right moment to ask: What does the day-to-day look like in the NYC lab versus the SF office? How is the GPU cluster accessed — on-prem, cloud, or hybrid? How often do the two hubs sync in person? The answers will reveal whether ATG's physical setup matches the intensity its research agenda demands.

The bi-coastal structure also creates a practical consideration for relocation. A candidate based in New York who joins the Quantitative Systems team could theoretically transfer to San Francisco later (or vice versa) without changing teams — the role exists in both locations. But the Research Scientist and Research Engineering roles appear NYC-locked, suggesting the deepest research leadership sits in the headquarters. If you're optimizing for long-term optionality across coasts, the platform and quantitative-systems tracks offer more geographic flexibility than the pure-research tracks.

Bottom line: ATG is a two-city company by design, not default. New York is the center of gravity for research and product leadership; San Francisco is a full peer for systems and platform engineering. Remote is a negotiated exception, not a policy. Candidates should decide which hub aligns with their discipline and whether they need the option to move between them — then verify the specifics in conversation with the team.

Who Thrives at Autonomous Technologies Group

The company describes itself in those same three phrases. That is not branding copy — it is a filter. ATG is an eight-person applied AI lab deploying frontier reasoning systems into financial markets, and its hiring signals make clear that the people who stay and ship share a specific profile.

First, agency over process. The job postings repeat "high agency" as a baseline expectation. In practice this means engineers and researchers who define their own problems, make architectural calls without waiting for a spec review, and treat ambiguity as the default state. The Data Platform MTS role makes this explicit: "You will take ownership of it end to end: hardening what exist today, making the core architectural calls on storage, orchestration, and data modeling, and building out the rest." There is no platform team to hand off to; the person who ingests the vendor feeds also designs the versioning scheme that prevents look-ahead bias for the research team.

Second, fluency in both AI systems and financial data. The stack is Python, SQL, AI agents, and a REST API added in September 2026, but the preferred background calls out "financial-data experience strongly preferred: the core domains cited above, or quantitative research." Candidates who have never wrestled with a corporate-action feed or a point-in-time survivorship-bias trap will spend months learning what the person next to them already knows. The Research Scientist role doubles down: "Drive original AI research focusing on new models and algorithms for reasoning in complex environments." Complex environments here means messy, non-stationary market data, not a clean benchmark suite.

Third, first-principles product thinking. The founders who sold Paperspace (a GPU cloud with a nine-figure exit, per DataGod's Wikipedia data), then "couldn't find a wealth manager they'd actually use." That origin story shapes the culture: the team is building the tool they wanted as customers. The Client Director role, one of two open sales-track positions, is not a traditional relationship manager. It owns "high-net-worth client relationships: source prospects, lead onboarding, drive adoption, define client service standards and workflows, build CRM and reporting systems, and relay customer insights to product and engineering." The last clause is the tell: sales feeds product directly, and the loop is tight.

Fourth, the AI-native workflow baseline. That posting states: "AI-native workflow: the posting notes the team does so daily to build data systems that are safe, well-described, and machine-consumable, for agents as much as humans." This is not aspirational; it is the current daily practice. Engineers who treat LLMs as autocomplete toys will slow down the team. The expectation is that you prompt, review, and integrate agent output as a standard part of the development cycle, and that you design schemas and APIs with agent consumers in mind.

Fifth, elite-team orientation over headcount growth. "We're building a small, elite team" appears in the YC listing. With eight people total and seven open roles split evenly between AI/ML and sales, the next hires will roughly double the company. Glassdoor shows only two reviews (61% would recommend, 4.1/5 work-life balance, 3.8/5 culture) — too few to be statistically meaningful, but consistent with a team too small for institutional review dynamics. The "stale roles: 100%" metric from alion.io (as of October 2026) likely reflects that the same seven requisitions have been open while the team interviews carefully rather than fills seats.

Sixth, disciplinary breadth with a shared language. The careers page echoes this, seeking curious minds from similarly diverse backgrounds. However, the actual roles — Research Scientist, Quantitative Systems, Data Platform, Backend, Research Engineering — all require enough systems engineering to ship production code. A pure theorist without engineering chops will not clear the bar; a pure engineer without research curiosity will not find the problems interesting.

The compensation band ($200k–$300k base for most roles, median $260k per first-party board data) reflects a market-rate Bay Area / NYC package for senior ICs, but the equity upside is tied to a pre-seed company that raised $15M led by Garry Tan and YC with BoxGroup and a top quant fund participating. The bet is binary: either the autonomous financial advisor becomes the default for high-net-worth allocation, or it doesn't. People who thrive here treat that binary as a feature, not a bug.

In short: you thrive at ATG if you have shipped production ML systems, you understand financial data plumbing, you default to writing code over writing tickets, you use coding agents as a force multiplier, and you want your next career chapter to be defined by a single, high-conviction product bet rather than a ladder of promotions.

The founders who sold Paperspace didn't hire a search firm to fill these seats. They wrote the job specs themselves, set the experience floors at the level of their own hardest problems, and left the postings live until the right people walked through the door. The seven roles still open are the seven conversations they're still waiting to have.

Working in robotics? Zero G Talent tracks the openings: see every open Autonomous Technologies Group role, browse robotics jobs, the companies hiring, and the people building the field.