Farsight AI
AI-driven workflow automation for finance and insurance, automating decks, models, and memos.
Website: https://www.farsight-ai.com/
Cover Block
PUBLIC
| Field | Value |
|---|---|
| Name | Farsight AI |
| Tagline | AI-driven workflow automation for finance and insurance, automating decks, models, and memos |
| Headquarters | New York |
| Founded | 2022 |
| Stage | Series A |
| Business Model | SaaS |
| Industry | Fintech |
| Technology | AI / Machine Learning |
| Geography | North America |
| Growth Profile | Venture Scale |
| Founding Team | Co-Founders (3+) |
| Funding Label | Series A |
| Total Disclosed | ~$16,000,000 |
Links
PUBLIC
- Website: https://www.farsight-ai.com/
- LinkedIn: https://www.linkedin.com/company/farsight-ai
- Crunchbase: https://www.crunchbase.com/organization/farsight-ai
Executive Summary
PUBLIC
Farsight AI is a New York-based software company building workflow automation for financial services and insurance, with an explicit focus on tasks (decks, models, memos) where AI hallucination risk has historically kept large language models out of production [LinkedIn, 2025]. The company was founded in 2022 by three MIT graduates, Samir Dutta, Noah Faro, and Kunal Tangri, who pair finance pedigree with applied machine-learning experience: Dutta worked at Evercore and General Atlantic, while Faro and Tangri held engineering roles at Amazon and Hugging Face respectively [SignalFire, 2025] [Crunchbase]. In June 2025, Farsight closed a $16M Series A from SignalFire and RRE Ventures, with the company reporting a tenfold revenue increase and a fivefold customer-base increase during 2024 [PRNewswire, June 2025] [Fintech Global, June 2025]. The product is positioned against a small but well-capitalized cohort of finance-native AI vendors, including Rogo and Hebbia, that are competing to become the default analyst-workflow layer inside banks, asset managers, and insurers. Two recent moves (the August 2025 acquisition of Presentable AI and a September 2025 integration with PitchBook) suggest a deliberate stack-up of presentation-generation and private-market data inputs [Farsight AI, August 2025] [Farsight AI, September 2025]. Over the next 12 to 18 months, the questions worth tracking are renewal behavior at named enterprise customers, gross margin durability as inference costs evolve, and whether the Presentable AI tuck-in materially expands ACV. The combination of a finance-trained CEO, applied-ML co-founders, and tier-one venture backing makes Farsight a credible entrant in a category where buyer trust and domain accuracy will likely matter more than raw model quality.
Data Accuracy: GREEN -- Confirmed by PRNewswire, SignalFire, Crunchbase, and Farsight's own press.
Taxonomy Snapshot
| Axis | Value |
|---|---|
| Stage | Series A |
| Business Model | SaaS |
| Industry / Vertical | Fintech (financial services and insurance workflows) |
| Technology Type | AI / Machine Learning (LLM-based) |
| Geography | North America (HQ New York) |
| Growth Profile | Venture Scale |
| Founding Team | Three co-founders, MIT graduates |
| Funding | $16M Series A disclosed (June 2025) |
Company Overview
PUBLIC
Farsight AI was founded in 2022 in New York by Samir Dutta, Noah Faro, and Kunal Tangri, three MIT graduates who converged on a shared thesis: that the most labor-intensive parts of financial services work (pitch decks, valuation models, investment memos, underwriting documents) were structurally underserved by horizontal AI tooling because of accuracy, branding, and security requirements specific to regulated finance [PRNewswire, June 2025] [LinkedIn, 2025]. Dutta, the CEO, brought transactional finance experience from Evercore and growth-investing experience from General Atlantic; Tangri (COO) had worked as a machine-learning engineer at Hugging Face; Faro (CTO) came from Amazon [SignalFire, 2025] [Crunchbase] [ZoomInfo].
The public milestone trail is short but dense. The company spent its first roughly two-and-a-half years in build mode and customer development with limited press visibility, then surfaced publicly in June 2025 with a $16M Series A led by SignalFire alongside RRE Ventures [PRNewswire, June 2025]. The funding announcement coincided with the disclosure of a 10x revenue increase and 5x customer-base growth during 2024 [Fintech Global, June 2025]. Two months later, in August 2025, Farsight acquired Presentable AI, framing the deal as the next phase of workflow automation for financial services [Farsight AI, August 2025]. In September 2025, the company announced a PitchBook integration to feed private capital market intelligence into its workflows [Farsight AI, September 2025].
The corporate story, in short, is one of a quietly-built early-stage company that picked an opinionated wedge (high-accuracy finance documents) and is now using a fresh Series A to assemble both the data inputs (PitchBook) and the output layer (Presentable AI) around its core engine.
Data Accuracy: GREEN -- Confirmed by PRNewswire, SignalFire, Crunchbase, and Farsight's own announcements.
Product and Technology
MIXED
Farsight's public product description centers on automating financial workflows (decks, models, memos) while preserving enterprise branding and security [Farsight AI]. The LinkedIn company page sharpens the positioning, noting a focus on "complex, nuanced use cases where risks of AI hallucination are particularly high" [LinkedIn, 2025] [PUBLIC]. Crunchbase characterizes the company as building workflow automation tools specifically for financial and insurance data processing [Crunchbase] [PUBLIC]. The two product surfaces that have been publicly described are document generation (presentations and memos that match a firm's templates) and structured analysis of financial filings such as 10-Ks and 10-Qs, the latter the subject of a benchmarking experiment co-author Tangri published on Medium evaluating multiple LLMs against finance-document question answering [Medium] [PUBLIC].
The August 2025 acquisition of Presentable AI extends the deck-generation surface, suggesting Farsight is internalizing the presentation layer rather than treating it as a downstream export [Farsight AI, August 2025] [PUBLIC]. The September 2025 PitchBook integration adds a structured private-markets data feed, which is meaningful for the investment banking and private-equity workflows where company comparables and transaction history drive most of the work product [Farsight AI, September 2025] [PUBLIC]. Together, these moves point toward a platform that ingests both internal client data and licensed third-party datasets, runs domain-tuned reasoning over them, and emits brand-compliant documents.
The underlying tech stack is not publicly documented in detail. The co-founders' applied-ML backgrounds at Hugging Face and Amazon, plus the Medium benchmarking work, suggest the team is model-agnostic and willing to evaluate multiple foundation models against finance-specific accuracy criteria rather than committing to a single provider (inferred from public technical writing and team backgrounds) [PRIVATE]. Engineering hires surfaced on LinkedIn (Andrei Dumitrescu, Kairi Sameshima, Orlson Chan, plus AI engineers Kent G. and Arina Ghanizadegan) indicate a small but growing applied-ML and software team [LinkedIn] [ZoomInfo] [PUBLIC].
Data Accuracy: YELLOW -- Product surfaces confirmed by company website, LinkedIn, and press; underlying model architecture is not publicly disclosed.
Market Research and Opportunity
PUBLIC
Finance back-office and middle-office automation is one of the most concentrated buyer segments for enterprise AI, because the work is document-heavy, the documents are structured, and the buyers have the budget. Farsight is selling into a segment where the willingness to pay per seat is unusually high and where regulatory pressure on auditability creates a moat against generic horizontal tools.
No confirmed third-party TAM figure for "AI for financial workflow automation" appears in the captured research, so any sizing here would be analogous rather than direct. What the cited material does establish is the buyer profile and the demand vector. SignalFire's investment note frames the opportunity as modernizing "finance's most critical workflows," pointing to investment banking pitch creation, asset-management memo writing, and insurance underwriting as the initial wedge categories [SignalFire, 2025]. RRE's investment write-up similarly frames the bet as the future of finance work, with the implicit comparable being the seat-based economics of incumbents like FactSet, Bloomberg Terminal, and PitchBook itself [RRE Ventures]. The PitchBook integration is notable here: rather than competing with a data incumbent, Farsight is plugging into one, which is an early signal of how the company plans to coexist with the existing data stack rather than replace it [Farsight AI, September 2025].
The demand drivers most visible in the public record are three. First, the maturing accuracy of LLMs on structured financial filings, an area Tangri's Medium benchmark explicitly investigated for 10-K and 10-Q question answering [Medium]. Second, regulatory and compliance pressure inside banks and insurers that creates demand for tooling with audit trails and brand-controlled outputs, which is precisely the gap Farsight describes on its homepage [Farsight AI]. Third, the cost pressure on junior analyst headcount across investment banking and asset management, which makes any tool that genuinely compresses deck and memo turnaround time a board-level conversation.
Adjacent and substitute markets include horizontal LLM platforms (OpenAI's enterprise tier, Anthropic Claude for Enterprise), document-generation platforms not specific to finance, and the in-house AI teams that bulge-bracket banks have begun assembling. Regulatory forces cut both ways: stricter model-risk-management guidance from US and European regulators raises the bar for any vendor selling into Tier 1 banks, but it also makes domain-specialized vendors with documented accuracy testing more attractive than general-purpose tools.
| Metric | Value | Period | Source |
|---|---|---|---|
| Revenue growth | 10x | 2024 | [Fintech Global, June 2025] |
| Customer-base growth | 5x | 2024 | [Fintech Global, June 2025] |
| Series A raised | $16M | June 2025 | [PRNewswire, June 2025] |
The disclosed growth multiples are striking but should be read with the usual early-stage caveat: a 10x revenue jump from a small base is a different signal than 10x from an established run-rate, and the absolute ARR is not public. The investor-grade read is that growth velocity was strong enough through 2024 to clear two tier-one VC diligence processes in 2025.
Data Accuracy: YELLOW -- Growth multiples confirmed by Fintech Global and PRNewswire; absolute ARR and TAM figures are not publicly disclosed.
Competitive Landscape
MIXED
Farsight is competing in a narrow but well-funded segment of finance-native AI assistants, with the principal named alternatives being Rogo and Hebbia.
| Company | Positioning | Stage / Funding | Notable Differentiator | Source |
|---|---|---|---|---|
| Farsight AI | Workflow automation for finance and insurance: decks, models, memos with brand and security controls | Series A, ~$16M disclosed | Focus on hallucination-sensitive workflows; PitchBook integration; Presentable AI acquisition | [PRNewswire, June 2025] [Farsight AI, September 2025] |
| Rogo | AI analyst tooling for investment banking and finance professionals | Later-stage venture funded | Direct sell into bulge-bracket and middle-market banking workflows | [PUBLIC, structured facts] |
| Hebbia | AI document analysis platform for financial services and professional services | Series B funded | Long-context document retrieval and analysis at large institutional scale | [PUBLIC, structured facts] |
The segment splits roughly into three competitive layers. The first layer is the finance-native AI cohort, where Farsight, Rogo, and Hebbia each pursue overlapping but distinct wedges: Hebbia is most associated with deep document analysis and retrieval at scale inside hedge funds and asset managers; Rogo has been visibly courting investment-banking analyst workflows; Farsight's public messaging emphasizes the document-output side (decks, models, memos) and explicitly extends into insurance, which neither competitor foregrounds publicly. The second layer is horizontal AI infrastructure (OpenAI Enterprise, Anthropic, Microsoft Copilot for Finance) that finance customers can use directly, often as a cheaper baseline. The third layer is the data and workflow incumbents (Bloomberg, FactSet, S&P Capital IQ, PitchBook) that own the underlying data feeds and could ship competing AI features inside their existing seats.
Where Farsight has a defensible edge today is the combination of a finance-trained CEO (Dutta's Evercore and General Atlantic experience reads as credible to buyers in the same seat) and an applied-ML bench from Hugging Face and Amazon, paired with the explicit positioning around hallucination-sensitive use cases [SignalFire, 2025] [LinkedIn, 2025]. The Presentable AI acquisition gives Farsight ownership of the output layer rather than dependency on a third-party generator, which matters in a category where brand-compliant decks are themselves the product [Farsight AI, August 2025]. That edge is meaningful but perishable: the underlying model capability gap between specialists and well-resourced horizontal platforms tends to compress over time, so the durable moat will likely come from workflow lock-in, proprietary fine-tuning data from customer engagements, and integrations like the PitchBook tie-up [Farsight AI, September 2025].
The most plausible source of exposure is incumbent distribution. PitchBook, Bloomberg, and FactSet sit on hundreds of thousands of finance seats already; an in-product AI feature from any of them, even a mediocre one, would force Farsight to compete on incremental value rather than first-mover novelty. A second exposure is the in-house build path at the largest banks, several of which have already stood up internal generative-AI teams. Farsight's counter is to be the best vendor on the narrow problem before incumbents notice it, and to land enough Tier 1 logos to make a build-versus-buy decision lopsided in its favor.
The 18-month scenario worth watching: Farsight wins if it can convert its 2024 growth velocity into two or three named, referenceable Tier 1 bank or insurer logos that publicly endorse the deck-and-memo workflow, which would compound into category-leader perception ahead of Rogo's and Hebbia's parallel pushes. Farsight is most exposed if PitchBook or Bloomberg ships a competitive AI assistant inside their existing terminal seats before Farsight has locked in renewal contracts.
Data Accuracy: YELLOW -- Competitor identities confirmed by structured research; relative funding stages and product depth are based on public characterization rather than disclosed customer counts.
Opportunity
PUBLIC
If Farsight executes on the wedge it has chosen, the prize is becoming the default document-and-analysis surface for a meaningful share of the global finance and insurance professional workforce.
The headline opportunity. The single largest outcome Farsight could plausibly become is the Bloomberg-Terminal-equivalent for AI-generated finance work product: the seat-based tool that an analyst, associate, or underwriter opens to draft a pitch, build a model, or write a memo, with the brand-compliance and accuracy guarantees their compliance team requires. The reason this outcome is reachable rather than aspirational is that the buyer category (investment banks, asset managers, insurers) already pays four-to-five-figure annual seat fees for finance-specific tools, the workflows being automated are document-shaped rather than judgment-shaped, and the company has already signaled willingness to integrate with rather than displace the existing data layer [Farsight AI, September 2025]. The 10x revenue and 5x customer growth disclosed for 2024 suggest that the wedge is finding paying buyers at meaningful velocity [Fintech Global, June 2025].
Growth scenarios.
| Scenario | What happens | Catalyst | Why it's plausible |
|---|---|---|---|
| Investment banking standard | Farsight becomes the default deck-and-memo tool inside two or three bulge-bracket or top middle-market banks | A named Tier 1 bank reference customer announces firm-wide deployment | Dutta's Evercore background gives credible distribution into the IB buyer; Presentable AI acquisition strengthens deck output [Farsight AI, August 2025] |
| Insurance underwriting expansion | Farsight extends from finance into insurance underwriting and claims memo automation | A named carrier deploys Farsight for underwriting workflow at scale | Company already positions for insurance on LinkedIn and Crunchbase, and the document patterns are analogous [LinkedIn, 2025] [Crunchbase] |
| Private-markets workflow layer | Farsight becomes the AI execution layer on top of private-capital data infrastructure | Deepening of the PitchBook integration into co-marketed product, plus similar tie-ups with Preqin or S&P | The September 2025 PitchBook integration is a real, public first step rather than a hypothetical [Farsight AI, September 2025] |
What compounding looks like. The flywheel that turns one win into the next is workflow data: every deck, model, and memo Farsight helps produce inside a customer becomes (with permission) feedback that improves the firm-specific fine-tuning, brand templates, and accuracy benchmarks for that customer, raising switching costs over each renewal cycle. A second compounding vector is integration density: each licensed data source (PitchBook so far) reduces the customer's reason to assemble the workflow themselves and creates a multi-vendor relationship that is harder to displace [Farsight AI, September 2025]. A third is hiring use: a finance-credible CEO plus an applied-ML bench attracts the kind of engineer-analyst hybrid talent that this category specifically rewards [SignalFire, 2025].
The size of the win. A useful (scenario, not a forecast) comparable is the seat-based finance-tooling category: Bloomberg L.P. is widely reported to generate over $10 billion in annual revenue from terminal seats, and FactSet trades as a public company with multibillion-dollar revenue. Even capturing a small single-digit percentage of the AI-augmented seat budget across investment banking, asset management, and insurance over the next decade would imply a multibillion-dollar revenue opportunity for whichever vendor establishes category default status. Farsight is one of three or four credibly-positioned contenders for that slot today, alongside Rogo and Hebbia. The path is narrow and the execution risk is real, but the prize is genuinely large, and the company has assembled the team, the capital, and the early integrations to compete for it.
Data Accuracy: YELLOW -- Scenarios are constructed from confirmed integrations, acquisitions, and team backgrounds; revenue comparables are public knowledge but applied here as analogies rather than forecasts.
Sources
PUBLIC
[Farsight AI] Farsight AI homepage | https://www.farsight-ai.com/
[Crunchbase] Farsight - Crunchbase Company Profile & Funding | https://www.crunchbase.com/organization/farsight-ai
[LinkedIn, 2025] Farsight AI company page | https://www.linkedin.com/company/farsight-ai
[PRNewswire, June 2025] Farsight Raises $16M in Funding, Announces Series A to Automate Financial Workflows and Decision-Making | https://www.prnewswire.com/news-releases/farsight-raises-16m-in-funding-announces-series-a-to-automate-financial-workflows-and-decision-making-302478363.html
[Bloomberg] Samir Dutta, Farsight AI Inc: Profile and Biography | https://www.bloomberg.com/profile/person/24915423
[LinkedIn] Kunal Tangri - Co-Founder at Farsight AI | https://www.linkedin.com/in/kunal-tangri-61ba48121/
[SignalFire, 2025] Farsight AI secures $16M Series A to modernize finance's most critical workflows | https://www.signalfire.com/blog/farsight-investor
[LinkedIn] David Yun - Farsight | https://www.linkedin.com/in/david-yun-9320a712b/
[Crunchbase] Kunal Tangri - Co-Founder at Farsight | https://www.crunchbase.com/person/kunal-tangri
[RRE Ventures] Our Investment in Farsight AI: The Future of Finance | https://blog.rre.com/our-investment-in-farsight-ai-the-future-of-finance-159de442c7f7
[Fintech Futures] AI workflow start-up Farsight raises $16m Series A | https://www.fintechfutures.com/venture-capital-funding/ai-workflow-start-up-farsight-raises-16m-series-a
[Fintech Global, June 2025] Coverage of Farsight Series A and 2024 growth metrics | https://www.fintech.global
[ZoomInfo] Farsight AI - Overview, News & Similar companies | https://www.zoominfo.com/c/farsight-ai/1335655757
[Medium] An LLM Benchmark for Financial Document Question Answering | https://medium.com/@kunal_79217/an-llm-benchmark-for-financial-document-question-answering-e63e9d2bda25
[Farsight AI] Farsight Raises $16M to Transform Finance with AI | https://www.farsight-ai.com/news/series-a-announcement
[Farsight AI, August 2025] Farsight Acquires Presentable AI | https://www.farsight-ai.com/news/farsight-acquires-presentable-ai
[Farsight AI, September 2025] Farsight integrates with PitchBook | https://www.farsight-ai.com/news
Articles about Farsight AI
- Farsight AI Is Putting Every Banker's Pitch Deck Through an LLM — The MIT-founded New York startup raised $16M from SignalFire and RRE to automate decks, models, and memos for finance and insurance teams.