AI agent observability and security — AI data 2026 | Stack Overflow Developer Survey

Developer Survey 2026

Question

Have you used any of the following tools for AI agent observability, prompts, or evals in the past year? Do you want to use any of the following in the next 6 months?

scale
Optional
v2026.1

Data

	Respondents	Percent
Sentry	592	26.0%
LangSmith	393	17.3%
MLflow	347	15.2%
Langfuse	345	15.2%
Datadog LLM	271	11.9%
Weights & Biases	244	10.7%
Arize AI	133	5.8%
Braintrust	132	5.8%
DeepEval	131	5.8%
Respan	127	5.6%
Promptfoo	121	5.3%
Agenta	114	5.0%
Ragas	111	4.9%
Galileo AI	110	4.8%
Portkey	109	4.8%
LangWatch	101	4.4%
Helicone	95	4.2%
Humanloop	93	4.1%
PromptLayer	92	4.0%
Patronus AI	91	4.0%
HoneyHive	90	4.0%
Confident AI	86	3.8%
Traceloop	86	3.8%
Chamber	85	3.7%
Opik	84	3.7%
Future AGI	80	3.5%
Sentrial	79	3.5%
Athina AI	78	3.4%
Lunary	77	3.4%
Moda	76	3.3%
Ashr	70	3.1%
Parea AI	69	3.0%

Use this data

Citation

Response data is released under the ODbL 1.0, which asks that you attribute it.

Stack Exchange Inc. (2026). “AI agent observability and security”. Stack Overflow Developer Survey 2026. Licensed under ODbL 1.0. https://survey.stackoverflow.co/2026/ai/data/ai-obs

JSON

Structured for code, with the question, the year and the source url alongside the numbers.

{
  "question": "AI agent observability and security",
  "dataId": "AIObs",
  "asked": "Have you used any of the following tools for AI agent observability, prompts, or evals in the past year? Do you want to use any of the following in the next 6 months?",
  "year": "2026",
  "url": "https://survey.stackoverflow.co/2026/ai/data/ai-obs",
  "demographic": {
    "id": "used",
    "type": "aiseries",
    "name": "Used",
    "n": 2277,
    "share": null
  },
  "series": null,
  "columns": [
    {
      "key": "response",
      "header": "AI agent observability and security",
      "unit": "",
      "numeric": false
    },
    {
      "key": "count",
      "header": "Respondents",
      "unit": "",
      "numeric": true
    },
    {
      "key": "pct",
      "header": "Percent",
      "unit": "%",
      "numeric": true
    }
  ],
  "licence": {
    "name": "ODbL 1.0",
    "url": "https://opendatacommons.org/licenses/odbl/1-0/",
    "holder": "Stack Exchange Inc."
  },
  "source": "Stack Overflow Developer Survey",
  "data": [
    {
      "response": "Sentry",
      "count": 592,
      "pct": 0.26
    },
    {
      "response": "LangSmith",
      "count": 393,
      "pct": 0.173
    },
    {
      "response": "MLflow",
      "count": 347,
      "pct": 0.152
    },
    {
      "response": "Langfuse",
      "count": 345,
      "pct": 0.152
    },
    {
      "response": "Datadog LLM",
      "count": 271,
      "pct": 0.119
    },
    {
      "response": "Weights &amp; Biases",
      "count": 244,
      "pct": 0.107
    },
    {
      "response": "Arize AI",
      "count": 133,
      "pct": 0.058
    },
    {
      "response": "Braintrust",
      "count": 132,
      "pct": 0.058
    },
    {
      "response": "DeepEval",
      "count": 131,
      "pct": 0.058
    },
    {
      "response": "Respan",
      "count": 127,
      "pct": 0.056
    },
    {
      "response": "Promptfoo",
      "count": 121,
      "pct": 0.053
    },
    {
      "response": "Agenta",
      "count": 114,
      "pct": 0.05
    },
    {
      "response": "Ragas",
      "count": 111,
      "pct": 0.049
    },
    {
      "response": "Galileo AI",
      "count": 110,
      "pct": 0.048
    },
    {
      "response": "Portkey",
      "count": 109,
      "pct": 0.048
    },
    {
      "response": "LangWatch",
      "count": 101,
      "pct": 0.044
    },
    {
      "response": "Helicone",
      "count": 95,
      "pct": 0.042
    },
    {
      "response": "Humanloop",
      "count": 93,
      "pct": 0.041
    },
    {
      "response": "PromptLayer",
      "count": 92,
      "pct": 0.04
    },
    {
      "response": "Patronus AI",
      "count": 91,
      "pct": 0.04
    },
    {
      "response": "HoneyHive",
      "count": 90,
      "pct": 0.04
    },
    {
      "response": "Confident AI",
      "count": 86,
      "pct": 0.038
    },
    {
      "response": "Traceloop",
      "count": 86,
      "pct": 0.038
    },
    {
      "response": "Chamber",
      "count": 85,
      "pct": 0.037
    },
    {
      "response": "Opik",
      "count": 84,
      "pct": 0.037
    },
    {
      "response": "Future AGI",
      "count": 80,
      "pct": 0.035
    },
    {
      "response": "Sentrial",
      "count": 79,
      "pct": 0.035
    },
    {
      "response": "Athina AI",
      "count": 78,
      "pct": 0.034
    },
    {
      "response": "Lunary",
      "count": 77,
      "pct": 0.034
    },
    {
      "response": "Moda",
      "count": 76,
      "pct": 0.033
    },
    {
      "response": "Ashr",
      "count": 70,
      "pct": 0.031
    },
    {
      "response": "Parea AI",
      "count": 69,
      "pct": 0.03
    }
  ]
}

Markdown

A formatted table, ready to paste into a document, an issue or a prompt.

|  | Respondents | Percent |
| --- | ---: | ---: |
| Sentry | 592 | 26.0% |
| LangSmith | 393 | 17.3% |
| MLflow | 347 | 15.2% |
| Langfuse | 345 | 15.2% |
| Datadog LLM | 271 | 11.9% |
| Weights &amp; Biases | 244 | 10.7% |
| Arize AI | 133 | 5.8% |
| Braintrust | 132 | 5.8% |
| DeepEval | 131 | 5.8% |
| Respan | 127 | 5.6% |
| Promptfoo | 121 | 5.3% |
| Agenta | 114 | 5.0% |
| Ragas | 111 | 4.9% |
| Galileo AI | 110 | 4.8% |
| Portkey | 109 | 4.8% |
| LangWatch | 101 | 4.4% |
| Helicone | 95 | 4.2% |
| Humanloop | 93 | 4.1% |
| PromptLayer | 92 | 4.0% |
| Patronus AI | 91 | 4.0% |
| HoneyHive | 90 | 4.0% |
| Confident AI | 86 | 3.8% |
| Traceloop | 86 | 3.8% |
| Chamber | 85 | 3.7% |
| Opik | 84 | 3.7% |
| Future AGI | 80 | 3.5% |
| Sentrial | 79 | 3.5% |
| Athina AI | 78 | 3.4% |
| Lunary | 77 | 3.4% |
| Moda | 76 | 3.3% |
| Ashr | 70 | 3.1% |
| Parea AI | 69 | 3.0% |

Spreadsheet

.csv is a plain text format which most spreadsheet software can open.