When Junior Smart founded the SOS Project in October 2006 to help young people leave gangs safely, he was met with suspicion. Many statutory bodies viewed as a risk a project that trains ex-gang members and offenders to mentor young people affected by gang culture.
But a decade on, the project, run by the London-based offender support charity the St Giles Trust, employs 27 people and works across 14 London boroughs.
The project was founded by Smart, a softly spoken Londoner, the day after he left prison having served five years for gang-related drug offences.
"Over time people have realised there is no one better positioned than ex-gang members to provide the right support to the young people most at risk," he says.
In the first year, not one of its 30 clients reoffended and SOS estimates that it has so far helped 3,000 young people to leave gangs successfully.
But the problems have intensified in the past decade, Smart says. Children as young as eight are being exploited as unpaid drug runners and carrying knives because they feel threatened.
When they're caught, hefty sentences and a lack of employment opportunities push them back into gangs on their release. The problem is also beginning to spread outside London, he says.
Ability to innovate fast
In an era of tight funding - 60 per cent of its money comes from the London Mayor's Office for Policing and Crime through local authorities, and the rest from voluntary income - Smart says one of the strengths of SOS has been its ability to innovate quickly. For example, the SOS+ service, which talks to schoolchildren about the realities of gang violence, was formed in just six weeks in 2008 after 14-year-old David Idowu was stabbed to death in the street.
It has also built strong relationships with the police, the probation and social services and the NHS through working at the Royal London Hospital's major trauma centre to identify and support victims of gang violence. But when asked what impact government policy has had on the project, Smart shrugs his shoulders. Policy interest is sporadic, he says, and usually driven by one case sparking a media outcry that quickly dies down.
Punitive drugs policies with deterrent sentences have no effect, he says, because young people don't believe they'll be caught.
And although London's schools have gone from being rated among the worst in the country in 2000 to among the best now, education policy has missed the most at-risk young people because many will drop out early, Smart says, or are "warehoused" in pupil referral units and become easy targets for gangs.
Smart is now cautiously turning his attention towards lobbying government to make policy changes he thinks would help. These include making it easier for employers to hire ex-offenders.
'What needs doing'
But he says: "I'm clever enough to know we can't just say 'this needs to be changed'
because the statutory bodies will come up with their own ideas. We need to be able to say what it is that needs doing."
Smart is currently working on a PhD on the issue and is hoping to use it to develop a ready-made policy solution. And, again, the SOS model is key.
"We've managed to give those with convictions real employment by taking their weaknesses and turning them into fantastic assets," he says. "But I want to see bigger changes.
"I always said if I could save one person, my time in prison would have been worth it."