Why Build Your Research Assistant with Bigdata.com?

Bigdata.com is designed specifically for business and finance professionals, delivering real-time, AI-driven insights with precision and speed. ✅ Unmatched Data Access — Instantly analyze vast, high-quality datasets.
✅ Finance & Business Expertise — Designed for professionals, not generic use.
✅ Seamless Integration — Easily power your chatbot with cutting-edge AI.
In just 5 minutes, you’ll learn how to: ✅ Create your API Key
✅ Craft a research prompt
✅ Chat with Bigdata.com Research Agent
✅ Advance filters: temporal, entity, sources
Ready to get started? Let’s dive in!

Choose research_effort equal to lite for a quick response, equivalent to the former Chat Service experience.

Create your API Key

Please click here to create your API Key: Developer Platform > API Keys

Craft a research prompt

The Research Agent accepts complex prompts, you can specify the role to the LLM, describe the research steps that you want to follow, and the output format you want. Example:

You are a Senior Portfolio Manager analyzing companies for institutional investors. 
Write a concise financial analysis of NVIDIA (NVDA) in Markdown. Include:
 1) Company overview
 2) Recent financial performance
 3) Key growth drivers
 4) Valuation vs peers
 5) Main risks, and
 6) Your investment outlook. 
 
 Use a professional, data-driven tone with first-person plural ('we believe'). 
 Structure as: # NVIDIA Financial Analysis ## Overview ## Financials ## Outlook ## Risks and Conclusion.

Chat with Bigdata.com Research Agent

The following script sends a single request to the Research Agent API using the Deep Research mode and starts printing the response as it comes. You can control the research effort with the parameter research_effort. It supports two values:

lite: Quick response, equivalent to the former Chat Service experience.
standard: The Research Agent will perform multiple steps until it has enough data to respond. It conducts Deep Research, and we recommend it for complex prompts.

import requests
import json
from dotenv import load_dotenv
import os
   
# Load environment variables from .env file
load_dotenv()
   
# Access the API key
my_api_key = os.getenv('BIGDATA_API_KEY')
endpoint = "https://agents.bigdata.com/v1/research-agent"

research_prompt = "You are a Senior Portfolio Manager analyzing companies for institutional investors. Write a concise financial analysis of NVIDIA (NVDA) in Markdown. Include: 1) Company overview, 2) Recent financial performance, 3) Key growth drivers, 4) Valuation vs peers, 5) Main risks, and 6) Your investment outlook. Use a professional, data-driven tone with first-person plural ('we believe'). Structure as: # NVIDIA Financial Analysis ## Overview ## Financials ## Outlook ## Risks and Conclusion."

payload = {
    "message": research_prompt,
    "research_effort": "standard"
}

headers = {
    "X-API-KEY": my_api_key,
    "Content-Type": "application/json",
}

# Stream and print research response in real-time
print("🔍 Starting research analysis...")

answer_text = ""
references = []

with requests.post(endpoint, headers=headers, json=payload, stream=True, timeout=60) as r:
    r.raise_for_status()
    for raw_line in r.iter_lines(decode_unicode=True):
        if not raw_line:
            continue
        if raw_line.startswith("data: "):
            data = raw_line[6:].strip()
            try:
                event = json.loads(data)
                message_type = event.get("message", {}).get("type")
                
                # Print ANSWER content as it comes and buffer it so grounding
                # offsets can be applied later.
                if message_type == "ANSWER":
                    content = event.get("message", {}).get("content", "")
                    if content:
                        answer_text += content
                        print(content, end="", flush=True)
                
                # Collect GROUNDING references for citation rendering. See
                # how-to-guides/agents/concepts/grounding-and-citations for the full
                # offset-mapping and rendering pattern.
                elif message_type == "GROUNDING":
                    references.extend(event.get("message", {}).get("references", []))
                
                # Signal completion
                elif message_type == "COMPLETE":
                    print(f"\n🎯 Research complete! Collected {len(references)} source references.")
                    
            except json.JSONDecodeError:
                # See how-to-guides/agents/concepts/error-handling for full error guidance.
                pass

Output:

# NVIDIA Financial Analysis

## Overview
NVIDIA Corporation (NASDAQ: NVDA) is a global leader in providing graphics, compute, and networking solutions. The company's core technologies revolve around its Graphics Processing Units (GPUs), which power diverse applications from high-end gaming to advanced artificial intelligence (AI) and data center workloads. NVIDIA operates primarily across several key segments: Data Center, Gaming, Professional Visualization, and Automotive . We believe NVIDIA holds a dominant market position, especially in the rapidly expanding AI accelerator space, commanding an estimated 80% to 90% share. Its technological prowess extends to software platforms like CUDA, which serves as a critical ecosystem for AI development.

## Financials
NVIDIA has demonstrated exceptional financial performance driven by the surge in AI demand.
*   **Revenue Growth:** In fiscal year 2025, NVIDIA reported revenue of $130.50 billion, marking an impressive 114.2% year-over-year (YoY) increase . The Data Center segment has been the primary growth engine, contributing $115.19 billion in FY25. More recently, Q2 FY26 revenue reached $46.74 billion, a 56% YoY increase, with the Data Center segment contributing $41.1 billion, also up 56% YoY . Gaming revenue in Q2 FY26 also saw a significant rebound, reaching $4.3 billion, a 49% increase from the prior year .
*   **Profitability:** The company maintains robust profitability, with a gross margin of 69.85% and a net margin of 52.41% . Operating income for FY25 stood at $81.45 billion, a 147.0% YoY increase, while net income for the same period was $72.88 billion, up 144.9% YoY .
*   **Earnings Per Share (EPS):** Diluted EPS for FY25 was $2.94, representing a 147.1% YoY growth . The latest reported EPS for Q2 FY26 was $1.05, surpassing analyst estimates .
*   **Cash Flow:** NVIDIA demonstrates strong cash generation, with operating cash flow reaching $64.09 billion in FY25, an increase of 128.2% YoY. Free cash flow also saw substantial growth, hitting $60.85 billion in FY25, up 125.2% YoY .

**Key Growth Drivers:**
NVIDIA's growth is primarily fueled by:
*   **Advancements in AI:** The company's dominance in AI accelerators, including the Blackwell architecture and the new Vera Rubin NVL144 CPX, positions it at the forefront of the AI revolution, catering to the massive demand from hyperscalers, neo-clouds, and enterprises .
*   **Data Center Expansion:** The global build-out of AI data centers heavily relies on NVIDIA's chips, with the company actively participating in large-scale data center projects (e.g., a $40 billion data center project in Texas) . NVIDIA is also leading the transition to more efficient 800-volt DC data center architectures .
*   **Gaming Market Trends:** Continuous innovation in gaming GPUs, such as the Blackwell-powered NVIDIA GeForce RTX 5060, and advancements in technologies like DLSS 4, maintain NVIDIA's strong position in the gaming sector .
*   **Professional Visualization and Automotive:** The adoption of RTX workstation GPUs for AI-powered design, simulation, and prototyping, alongside growth in automotive and robotics solutions like the DRIVE Hyperion sensor architecture, contributes to diversification and future growth .
*   **CUDA Software Ecosystem:** The robust and widely adopted CUDA software platform creates a significant competitive moat, ensuring customer lock-in and a persistent advantage in AI development .

## Outlook
To assess NVIDIA's valuation, we compare it against key competitors in the semiconductor industry: Advanced Micro Devices (AMD), Intel (INTC), and Broadcom (AVGO).

| Metric               | NVIDIA (NVDA) | AMD (AMD) | Intel (INTC) | Broadcom (AVGO) |
| :------------------- | :------------ | :-------- | :----------- | :-------------- |
| Market Cap (Billion) | $4,460.86     | $378.25   | $161.99      | $1,641.23       |
| P/E Ratio            | 51.55x        | 133.48x   | -7.89x       | 86.98x          |
| P/S Ratio            | 27.00x        | 12.78x    | 3.05x        | 27.39x          |
| EV/EBITDA            | 44.73x        | 67.98x    | 182.65x      | 53.34x          |
| Debt/Equity Ratio    | 0.11x         | 0.07x     | 0.52x        | 0.88x           |
| Gross Margin         | 69.85%        | 44.24%    | 29.76%       | 66.83%          |
| Net Margin           | 52.41%        | 9.57%     | -38.64%      | 31.59%          |
, , , 

NVIDIA trades at a higher Price-to-Earnings (P/E) ratio than Intel but lower than AMD and Broadcom. Its Price-to-Sales (P/S) ratio is comparable to Broadcom and significantly higher than AMD and Intel, reflecting its premium valuation driven by robust growth. Similarly, NVIDIA's Enterprise Value to EBITDA (EV/EBITDA) is elevated, though lower than Intel's and AMD's, suggesting that while its earnings are strong, its enterprise value relative to earnings before interest, taxes, depreciation, and amortization is still considered high within the sector. We note Intel's negative P/E ratio is due to its negative net income in the last fiscal year. NVIDIA's debt-to-equity ratio is low, indicating a strong balance sheet compared to its peers. The company's superior gross and net margins underscore its strong pricing power and cost efficiency.

We believe NVIDIA's investment outlook remains largely positive, underpinned by its irreplaceable role in the AI infrastructure build-out and its robust financial performance. The company's technological leadership, particularly with its new Blackwell and Rubin architectures, is likely to sustain demand for its products. However, the current valuation reflects significant future growth expectations, necessitating continued execution and innovation to justify its premium.

## Risks and Conclusion
While NVIDIA's prospects are compelling, several risks could impact its future performance:
*   **Competitive Pressures:** Competition is intensifying, with rivals like AMD introducing competitive GPUs (e.g., MI300X) and forging partnerships (e.g., with OpenAI) to challenge NVIDIA's market share . Broadcom is also making strides with custom AI chips .
*   **Geopolitical Factors:** Geopolitical tensions, particularly between the U.S. and China, pose a significant threat. Export restrictions have already impacted NVIDIA's sales to China, with the CEO noting a substantial decrease in market share in the region .
*   **Supply Chain Disruptions:** NVIDIA's reliance on third-party manufacturers, such as TSMC, exposes it to potential supply chain vulnerabilities. While the company is actively mitigating this through strategic investments (e.g., in Intel for advanced packaging), disruptions could still affect production and delivery timelines .
*   **Market Cyclicality and Demand Slowdown:** The high expectations surrounding the AI boom could lead to market cyclicality. There is a risk that the current rapid pace of AI spending could decelerate, or a shift in the AI lifecycle (e.g., from training to inference workloads) could impact demand for NVIDIA's high-end chips, potentially leading to an "overbuilt cycle" similar to past tech booms .
*   **Customer Concentration Risk:** A significant portion of NVIDIA's revenue is derived from a few large hyperscaler clients. Any changes in their procurement strategies or increasing development of custom silicon could pose a risk .
*   **Regulatory Changes:** Evolving regulatory landscapes and changes in industry standards could necessitate significant adaptations and investments.

In conclusion, we believe NVIDIA is a fundamentally strong company with an unparalleled position in the burgeoning AI market. Its financial performance has been exceptional, driven by technological leadership and a robust ecosystem. However, investors must be mindful of the premium valuation and the inherent risks associated with increasing competition, geopolitical uncertainties, and potential market cyclicality. NVIDIA's ability to maintain its innovation edge and navigate these challenges will be crucial for sustaining its growth trajectory.
🎯 Research complete!

Next steps

The example above only handles ANSWER, GROUNDING, and COMPLETE events. A production integration should also handle planning, tool calls, retries, and errors. The Concepts guides cover everything the stream can emit and how to respond to each.

Streaming responses

The 11 message types, their fields, and a complete dispatching handler.

Grounding and citations

Render the references you just collected as inline Markdown citations.

Error handling

HTTP-level errors, in-stream ERROR vs TOOL_ERROR vs LLM_RETRY, and retry strategy.

Conversation continuity

Continue Research Agent chats with chat_id and checkpoint_id.

Non-streaming print

Buffer the stream and print the complete answer in one go.

Monitor usage

See a breakdown of your organization’s usage per service.

Getting Started

Research Service

Search Service

Upload proprietary content

Knowledge Graph

Watchlist

Troubleshooting

Research Agent Quickstart Guide

Why Build Your Research Assistant with Bigdata.com?

Create your API Key

Craft a research prompt

Chat with Bigdata.com Research Agent

Next steps

Streaming responses

Grounding and citations

Error handling

Conversation continuity

Non-streaming print

Monitor usage

​Why Build Your Research Assistant with Bigdata.com?

​Create your API Key

​Craft a research prompt

​Chat with Bigdata.com Research Agent

​Next steps

Streaming responses

Grounding and citations

Error handling

Conversation continuity

Non-streaming print

Monitor usage

Why Build Your Research Assistant with Bigdata.com?

Create your API Key

Craft a research prompt

Chat with Bigdata.com Research Agent

Next steps