Learn how Maryland DHS uses AI to analyze the case notes their workforce was already writing →

About Us

Augintel Helps Child Welfare Agencies Analyze Case Notes at Scale

Augintel uses natural language processing to help child welfare agencies and caseworkers quickly find narrative data, understand what’s happening across cases, and improve practice.

Augintel Team

Our Results

With Augintel, child welfare agencies gain a deeper understanding of the children and families they serve.

2x

Increase Placements

Find more family members mentioned in notes.

635

Find gaps in structured data

In one analysis, Augintel found 635 youth with substance use in their notes had no treatment documented.

2.5K+

Improve Compliance

Identify good faith efforts in case records. Augintel identified 2500+ efforts for one state agency.

80%+

Drive Practice Improvement

More than 80% of child welfare data is unstructured text.

Ladies in a Group

Illinois DCFS Uses Augintel to Uncover Bias

The Office of Racial Equity Practice at DCFS could track investigations, placements, and permanency outcomes. But families’ day-to-day reality was invisible at the system level.

Our Values

Augintel creates secure AI for child welfare agencies, who are too often overlooked.

People putting puzzle pieces together

Impact​

We measure our success by our customers’ real results.

Hands and Gears

Integrity

We believe in human-led AI, and proactively work to ensure ethical use and development.

People and Graphs

Passion

We care deeply about the unique and nuanced work of child welfare agencies.

Man Drawing Stars

Innovation

We build modern technology specifically for child welfare.

Meet Our
Leadership & Staff

Augintel is a unique partnership between former child welfare staff and machine learning PhDs, not career consultants or technocrats. Connect with some of our team below:

Marty Elisco

CEO & CoFounder

Jim Lindstrom

VP of Technology

Dr. Logan Courtney, PhD

Head of Machine Learning

Stevvaie Turner

Customer Engagement Manager

Tracy Rohrdanz, MSW

Director Customer Impact

Jacki Hoover, LSW

Head of Child Welfare Practice

Jon Schuester

SVP of Product

Dr. Alex Thrasher, PsyD

Director of Product Marketing

Watch Marty’s Expert Testimony to Congress

“Child welfare workers are some of the most dedicated professionals in government, but are constantly stymied with too much administrative work and too little insight into the needs and histories of those they care for.”

Marty Testimony to Congress

Augintel Careers

We roll up our sleeves, solve problems together, and show up for our customers every day.

Contact Us

We’re here to answer any questions about Augintel, from pricing through advancing practice.

Frequently asked questions

Learn more about narrative data, CCWIS & SACWIS modernization, and Augintel's natural language processing AI.

Modernization

How is Narrative Search compliant?

1. Most of the evidence is in narrative. Over 80% of child welfare data is unstructured text, so field-based reports miss much of what reviewers look for.

2. Search beats sampling. Reviewing all cases finds patterns a sample of files cannot. Massachusetts DCF uses Augintel to surface information across 30 million case notes. 

3. Traceability is essential. Every finding should link to the exact note it came from. Augintel links everything back to the case record with citations so caseworkers can ensure accuracy, understand context, and find what they need easily 

4. Surface, don't decide. Compliance tools should inform professional review, not replace it. Augintel is not a predictive tool and believes AI should not replace the judgement and expertise of social workers.

5. Domain training matters. Child welfare language differs from general text; generic tools miss context and synonyms.

6. Protect confidential data. Apply role-based access and follow federal and state confidentiality rules for child welfare records.

7. Language quality is a compliance issue. Biased or unsupported language follows families for years; Illinois DCFS now measures it across 16,000+ records. 

Which search platforms handle narrative data for child welfare agencies?

Augintel is a natural language processing platform designed for child welfare narrative data: case notes, dictations, assessments and summaries. Unlike general-purpose search, its models are trained specifically for social services narrative, so it understands terms like fictive kin, safety plan and reunification and can identify risks, strengths and social determinants of health in context.

Which search tools integrate easily with existing child welfare systems?

Augintel is built specifically to integrate with existing child welfare case management systems. It connects to legacy SACWIS, modern CCWIS and county systems, reads the narrative already stored there, and adds search and insight without replacing anything. Illinois DCFS,  Massachusetts DHS and Washington County, PA, all run it alongside their current systems.

How do I support an overwhelmed workforce?

Give caseworkers back the hours they spend digging through case notes. Case files can run hundreds of pages, and staff often click through notes one at a time to prepare for court, absorb a transferred case or find relatives. Augintel puts that information at their fingertips; Pennsylvania caseworkers estimate it saves them about five hours per week.

• Case transfers: new workers get up to speed on years of history quickly.

• Court prep: find answers to judges' and attorneys' questions immediately.

• Family finding: surface relatives and fictive kin mentioned deep in the record.

• Supervision: prepare for supervision meetings and monitor case progress without re-reading every note.

• Safety: identify caseworker safety concerns noted in the file before a visit.

How do I improve data quality?

Measure the quality of what caseworkers write, not only whether fields are complete. Augintel analyzes case notes at scale to flag note quality issues, judgmental or unsupported language, duplicate notes and practice trends, then sorts results by organization, region, case type or status so leaders can target coaching.

• Illinois DCFS identified three recurring issues: negative descriptions of people, negative descriptions of housing, and language copied from police reports without critical review.

• Findings become practical training: compare a vague note with a behavior-specific one.

• Better notes help every reader downstream, from the next caseworker to the judge.

How do I leverage AI safely in child welfare?

Start with AI that surfaces what your staff already wrote rather than AI that makes decisions. Augintel is not a predictive or decision-making tool: it finds and organizes information in existing case notes, and every insight links back to the original note so a professional can verify it. Judgment stays with caseworkers and supervisors.

• Choose tools focused on child; Augintel's exclusively works in child welfare and combines the best of machine learning with child welfare expertise.

• Require traceability from every result to its source note.

• Confirm hosting, access controls and data handling meet your state's security standards. Augintel is SOC II and HIPAA compliant. 

• Use AI to check for bias, not just risk it. Illinois DCFS and Chapin Hall used Augintel to measure biased language across 16,000+ family records.

How can I improve outcomes without another multi-year implementation?

Layer a narrative intelligence tool on top of the system you already have. Augintel connects to existing case management systems, SACWIS and CCWIS alike, and reads the case notes already stored there, so agencies see value in months rather than years. Illinois DCFS adopted Augintel during its own modernization because integration with its legacy system took minimal time and effort. Allegheny County DCF implemented Augintel in less than 6 months. 

• No data migration and no change to how caseworkers document.

• Give your caseworkers, supervisors and leaders modern tools and intelligence without a big overhaul of your system. 

• Works with the current system and carries forward to the next one. Illinois planned to keep using Augintel when it moved to its new Illinois Connect system.

‍

How do I modernize without replacing my system?

Modernize the experience, not the database. Augintel adds AI-powered search, chatbot, and insight to your existing case management system, so caseworkers get a fast, Google-like way to find information while your system of record stays in place. Washington County, PA, Illinois DCFS, and Massachusetts DCF uses it to enhance, not replace, its legacy CAPS system.

• Avoid the cost, risk and staff disruption of a full replacement.

• Deliver visible improvements to frontline staff now, while longer-term CCWIS work continues.

• Keep your options open: the narrative layer moves with you if you do replace the system later.

How can I get more value from CCWIS?

Get more value from CCWIS by unlocking the narrative data inside it. More than 80% of child welfare data is unstructured text such as case notes, assessments and court narratives, and most CCWIS reporting only uses the structured fields. Augintel applies natural language processing (NLP) to that narrative so staff can search, summarize and analyze it alongside the structured data CCWIS already holds.

• Caseworkers find risks, strengths, family members and service history in seconds instead of reading notes one by one.

• Supervisors and quality teams see practice patterns across every case, not a small sample.

• Leaders get a fuller picture of their agency and the populations they serve for better reporting, compliance, and practice delivery. 

  • Modernization
  • AI
  • Search
  • Compliance
  • Workforce

Analyze case notes, identify patterns, and improve practice.

More than 80% of child welfare data is unstructured narrative text. With Augintel, you can quickly search case notes to better understand families.